National Academies Press: OpenBook

Feasibility of Using In-Vehicle Video Data to Explore How to Modify Driver Behavior That Causes Nonrecurring Congestion (2011)

Chapter: Chapter 8 - Conclusions and Recommendations for Future Data Collection Efforts

« Previous: Chapter 7 - Potential Problems and Issues in Data Reduction
Page 64
Suggested Citation:"Chapter 8 - Conclusions and Recommendations for Future Data Collection Efforts." National Academies of Sciences, Engineering, and Medicine. 2011. Feasibility of Using In-Vehicle Video Data to Explore How to Modify Driver Behavior That Causes Nonrecurring Congestion. Washington, DC: The National Academies Press. doi: 10.17226/14509.
×
Page 64
Page 65
Suggested Citation:"Chapter 8 - Conclusions and Recommendations for Future Data Collection Efforts." National Academies of Sciences, Engineering, and Medicine. 2011. Feasibility of Using In-Vehicle Video Data to Explore How to Modify Driver Behavior That Causes Nonrecurring Congestion. Washington, DC: The National Academies Press. doi: 10.17226/14509.
×
Page 65
Page 66
Suggested Citation:"Chapter 8 - Conclusions and Recommendations for Future Data Collection Efforts." National Academies of Sciences, Engineering, and Medicine. 2011. Feasibility of Using In-Vehicle Video Data to Explore How to Modify Driver Behavior That Causes Nonrecurring Congestion. Washington, DC: The National Academies Press. doi: 10.17226/14509.
×
Page 66
Page 67
Suggested Citation:"Chapter 8 - Conclusions and Recommendations for Future Data Collection Efforts." National Academies of Sciences, Engineering, and Medicine. 2011. Feasibility of Using In-Vehicle Video Data to Explore How to Modify Driver Behavior That Causes Nonrecurring Congestion. Washington, DC: The National Academies Press. doi: 10.17226/14509.
×
Page 67
Page 68
Suggested Citation:"Chapter 8 - Conclusions and Recommendations for Future Data Collection Efforts." National Academies of Sciences, Engineering, and Medicine. 2011. Feasibility of Using In-Vehicle Video Data to Explore How to Modify Driver Behavior That Causes Nonrecurring Congestion. Washington, DC: The National Academies Press. doi: 10.17226/14509.
×
Page 68
Page 69
Suggested Citation:"Chapter 8 - Conclusions and Recommendations for Future Data Collection Efforts." National Academies of Sciences, Engineering, and Medicine. 2011. Feasibility of Using In-Vehicle Video Data to Explore How to Modify Driver Behavior That Causes Nonrecurring Congestion. Washington, DC: The National Academies Press. doi: 10.17226/14509.
×
Page 69
Page 70
Suggested Citation:"Chapter 8 - Conclusions and Recommendations for Future Data Collection Efforts." National Academies of Sciences, Engineering, and Medicine. 2011. Feasibility of Using In-Vehicle Video Data to Explore How to Modify Driver Behavior That Causes Nonrecurring Congestion. Washington, DC: The National Academies Press. doi: 10.17226/14509.
×
Page 70
Page 71
Suggested Citation:"Chapter 8 - Conclusions and Recommendations for Future Data Collection Efforts." National Academies of Sciences, Engineering, and Medicine. 2011. Feasibility of Using In-Vehicle Video Data to Explore How to Modify Driver Behavior That Causes Nonrecurring Congestion. Washington, DC: The National Academies Press. doi: 10.17226/14509.
×
Page 71
Page 72
Suggested Citation:"Chapter 8 - Conclusions and Recommendations for Future Data Collection Efforts." National Academies of Sciences, Engineering, and Medicine. 2011. Feasibility of Using In-Vehicle Video Data to Explore How to Modify Driver Behavior That Causes Nonrecurring Congestion. Washington, DC: The National Academies Press. doi: 10.17226/14509.
×
Page 72
Page 73
Suggested Citation:"Chapter 8 - Conclusions and Recommendations for Future Data Collection Efforts." National Academies of Sciences, Engineering, and Medicine. 2011. Feasibility of Using In-Vehicle Video Data to Explore How to Modify Driver Behavior That Causes Nonrecurring Congestion. Washington, DC: The National Academies Press. doi: 10.17226/14509.
×
Page 73
Page 74
Suggested Citation:"Chapter 8 - Conclusions and Recommendations for Future Data Collection Efforts." National Academies of Sciences, Engineering, and Medicine. 2011. Feasibility of Using In-Vehicle Video Data to Explore How to Modify Driver Behavior That Causes Nonrecurring Congestion. Washington, DC: The National Academies Press. doi: 10.17226/14509.
×
Page 74
Page 75
Suggested Citation:"Chapter 8 - Conclusions and Recommendations for Future Data Collection Efforts." National Academies of Sciences, Engineering, and Medicine. 2011. Feasibility of Using In-Vehicle Video Data to Explore How to Modify Driver Behavior That Causes Nonrecurring Congestion. Washington, DC: The National Academies Press. doi: 10.17226/14509.
×
Page 75
Page 76
Suggested Citation:"Chapter 8 - Conclusions and Recommendations for Future Data Collection Efforts." National Academies of Sciences, Engineering, and Medicine. 2011. Feasibility of Using In-Vehicle Video Data to Explore How to Modify Driver Behavior That Causes Nonrecurring Congestion. Washington, DC: The National Academies Press. doi: 10.17226/14509.
×
Page 76
Page 77
Suggested Citation:"Chapter 8 - Conclusions and Recommendations for Future Data Collection Efforts." National Academies of Sciences, Engineering, and Medicine. 2011. Feasibility of Using In-Vehicle Video Data to Explore How to Modify Driver Behavior That Causes Nonrecurring Congestion. Washington, DC: The National Academies Press. doi: 10.17226/14509.
×
Page 77
Page 78
Suggested Citation:"Chapter 8 - Conclusions and Recommendations for Future Data Collection Efforts." National Academies of Sciences, Engineering, and Medicine. 2011. Feasibility of Using In-Vehicle Video Data to Explore How to Modify Driver Behavior That Causes Nonrecurring Congestion. Washington, DC: The National Academies Press. doi: 10.17226/14509.
×
Page 78

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

C H A P T E R 8 Conclusions and Recommendations for Future Data Collection EffortsThe team reviewed reduced data from each candidate data set in Chapter 4. The original reduced data were not detailed or specific enough to recognize contributing factors to safety- related events. Consequently, additional data reduction was needed. The next section details the factors that contribute to crashes and near crashes, whether or not driver behavior can be adjusted, and corresponding countermeasures are recommended. Contributing Factors and Correctable Driver Behaviors For Project 5, RDCWS FOT, the original data reduction did not specifically pinpoint contributing factors to events. The variables coded for drivers and the related environment stated the situation only at the instant the events occurred. The team performed additional data reduction and identified contribut- ing factors. As seen in Table 8.1, in most of the safety-related events, driver-related decision errors are the contributing factor. For both freeway and arterial safety-related events, more than 80% of the cases are caused by factors in this category. The next largest category on both types of roads is recognition errors in which drivers were distracted or failed to look. For Project 6, the 100-Car Study, the factors that precipitated a safety-related event, contributed to the event, and were associated with the event were determined. These factors are grouped into pre-event maneuvers, precipitating factors, con- tributing factors, associated factors, and avoidance maneuvers. Of all the factors, contributing factors are the key in the study of driver behavior and were judged by trained data reductionists to have directly influenced the presence or severity of a crash or near crash. Three subcategories were further constructed for contributing factors to identify the causes of crashes: infrastructure factors and driving environment factors, such as road surface, traffic density, and weather; driver factors, such as driver inattention, drowsiness, and distraction; and vehicle factors, such as flat tires and vehicle breakdowns.64As revealed by the data, the factors that contributed to a crash usually were not caused by a sole factor but involved some form of factor interaction. For example, the driver could be distracted by both talking on a cell phone and adjusting the radio during an event, or the crash could have been caused by both inattention of the driver and a poorly designed roadway. Therefore, as shown in Tables 8.2 and 8.3, the resulting sum of percentages of contributing factors may add up to more than 100%. Not all 18 conflict categories are listed in these tables. Most of the categories of events have driver factors involved as a contributing factor, especially for the single- and lead-vehicle crashes, as well as for the single-, lead-, and following-vehicle near crashes. These categories have more than 100% driver factor–involved cases, demonstrating a high probability of human errors. In some cases the contributing factors are descriptions of the status of the driver, vehicle, and environment at the moment the event happened. For example, if the driver was using a wireless device when the event happened, the driver factor would be “Secondary task” under the “Inattention to Forward Roadway” category. It is possible that these factors are not the direct causal factor leading to the events. To better distinguish preventable events from others, additional data reduction was performed by the team. Table 8.4 ascribes the crashes and near crashes to one mutually exclusive contribut- ing category that can be considered as a dominant factor. The categories listed in this table better serve the purpose of studying the relationship between behavior and travel time reliability. In summary, the data collected in the 100-Car Study were comprehensively and accurately reduced. The continuous video and other data are suitable for studying driving behavior and its impact on travel time reliability. For Project 7, DDWS FOT, more than one vehicle was involved in multiple crashes. Because only the subject vehicle was equipped with data collection equipment, data reduc- tionists could observe only scenarios related to that vehicle. Contributing factors were analyzed based on the observations

65Table 8.1. Critical Contributing Factors for Project 5 Critical Factor Freeway Arterial Total Driver-related factor (critical nonperformance errors), including sleep; heart attack or other 0 (0.0%) 0 (0.0%) 0 (0.0%) physical impairment of the ability to act; drowsiness, fatigue, or other reduced alertness; other critical nonperformance Driver-related factor (recognition errors), including inattention; internal distraction; external 5 (6.5%) 0 (0.0%) 5 (5.0%) distraction; inadequate surveillance (e.g., failure to look); other or unknown recognition error Driver-related factor (decision errors), including too fast or too slow; misjudgment of gap; 66 (85.7%) 19 (82.6%) 85 (85.0%) following too closely or unable to respond to unexpected actions; false assumption of other road user’s actions; apparently intentional sign or signal violation; illegal U-turn or other illegal maneuver; failure to turn on headlamps; inadequate evasive action; aggressive driving; other or unknown decision error Driver-related factor (performance errors), including panic or freezing; overcompensation; 0 (0.0%) 0 (0.0%) 0 (0.0%) poor directional control; other or unknown performance error Environment-related factor, including sign missing; view obstruction by roadway design; 0 (0.0%) 0 (0.0%) 0 (0.0%) roadway geometry; sight distance; maintenance problems; slick roads; other highway-related conditions Environment-related factor, including glare, blowing debris, animal or object in roadway 0 (0.0%) 0 (0.0%) 0 (0.0%) Crashes or near crashes caused by others 1 (1.3%) 1 (4.3%) 2 (2.0%) Unknown reasons 5 (6.5%) 3 (13.0%) 8 (8/0%) Total 77 (100%) 23 (100%) 100 (100%)Table 8.2. Contributing Factors for Crashes in Project 6 Factor Category Crashes Total Number Driver Environmental Vehicle Single vehicle 24 121% 38% 0% Lead vehicle 15 127% 13% 0% Following vehicle 12 83% 8% 0% Object obstacle 9 144% 56% 0% Parked vehicle 4 100% 50% 0% Animal 2 0% 100% 0% Turning across opposite direction 2 100% 50% 0% Adjacent vehicle 1 100% 0% 0%from equipped vehicles. The most frequent critical reason for crashes was “object in roadway,” which constituted 57% of the total events. The next largest groups were driver-related factors (recognition errors) and driver-related factors (performance errors); each group had more than 14% of the cases. For tire strike cases, most were attributed to environment-related factors. For near crashes, driver-related factors (recognition errors and decision errors) constituted nearly half of all cases. Details of the critical factors are enumerated in Table 8.5. In summary, the data collected in this study were com- prehensive, and the data reduction was extensive. This studywas conducted more recently than Project 6; consequently, the instrumentation used to collect the data was more accurate. Because only commercial trucks were studied, the data set has certain limitations with regard to the versatility of drivers and vehicles. For Project 8, NTDS, a total of 320,011 triggers were visu- ally inspected during data reduction. From those triggers, 2,899 safety-critical events were identified, including 13 crashes (eight of those were tire strikes), 61 near crashes, 1,594 crash- relevant conflicts, 1,215 unintentional lane deviations, and 16 illegal maneuvers. Additionally, a random sample of

66Table 8.3. Contributing Factors for Near Crashes in Project 6 Factor Category Near Crashes Total Number Driver Environmental Vehicle Single vehicle 48 135% 29% 0% Lead vehicle 380 119% 9% 0% Following vehicle 70 110% 9% 0% Object obstacle 6 83% 50% 0% Parked vehicle 5 60% 0% 0% Animal 10 70% 10% 0% Turning across opposite direction 27 96% 30% 0% Adjacent vehicle 115 90% 11% 0% Merging vehicle 6 33% 67% 0% Across path through intersection 27 89% 30% 0% Oncoming 27 96% 30% 0% Other 2 100% 50% 0% Pedestrian 6 133% 50% 0% Turning across in same direction 3 44% 11% 0% Turning in same direction 28 54% 21% 0% Unknown 1 200% 0% 0%Table 8.4. Critical Factors Contributing to Crashes and Near Crashes Critical Factor Crashes Near Crashes Driver-related factor (critical nonperformance errors), including sleep; heart attack or other physical impairment of the ability to act; drowsiness, fatigue, or other reduced alertness; other critical nonperformance 8 (11.6%) 33 (4.3%) Driver-related factor (recognition errors), including inattention; internal distraction; external distraction; inadequate surveillance (e.g., failure to look); other or unknown recognition error 22 (31.9%) 201 (26.4%) Driver-related factor (decision errors), including too fast or too slow; misjudgment of gap; following too closely or unable to respond to unexpected actions; false assumption of other road user’s actions; apparently intentional sign or signal violation; illegal U-turn or other illegal maneuver; failure to turn on headlamps; inadequate evasive action; aggressive driving; other or unknown decision error 19 (27.5%) 218 (28.6%) Driver-related factor (performance errors), including panic or freezing; overcompensation; poor directional control; other or unknown performance error 1 (1.4%) 30 (3.9%) Environment-related factor, including sign missing; view obstruction by roadway design; roadway geometry; sight distance; maintenance problems; slick roads; other highway-related conditions 2 (2.9%) 26 (3.4%) Environment-related factor, including glare, blowing debris, animal or object in roadway 4 (5.8%) 29 (3.8%) Crashes or near crashes caused by others 13 (18.8%) 224 (29.4%) Total 69 (100%) 761 (100%)

67Table 8.5. Categorized Critical Factors for Crashes and Near Crashes in Project 7 Critical Factor Crashes Crashes: Tire Strikes Near Crashes Critical reason not coded to this vehicle. 1 7.1% 0 0% 29 29.6% Driver-related factor (critical nonperformance errors), including sleep; heart attack 0 0% 0 0% 1 1.0% or other physical impairment of the ability to act; drowsiness, fatigue, or other reduced alertness; other critical nonperformance Driver-related factor (recognition errors), including inattention; internal distraction; 2 14.3% 0 0% 30 30.6% external distraction; inadequate surveillance (e.g., failure to look); other or unknown recognition error Driver-related factor (decision errors), including too fast or too slow; misjudgment 1 7.1% 2 14.3% 18 18.4% of gap; following too closely or unable to respond to unexpected actions; false assumption of other road user’s actions; apparently intentional sign or signal violation; illegal U-turn or other illegal maneuver; failure to turn on headlamps; inadequate evasive action; aggressive driving; other or unknown decision error Driver-related factor (performance errors), including panic or freezing; over- 2 14.3% 3 21.4% 7 7.1% compensation; poor directional control; other or unknown performance error Environment-related factor, including sign missing; view obstruction by roadway 0 0% 9 64.3% 2 2.1% design; roadway geometry; sight distance; maintenance problems; slick roads; other highway-related conditions Environment-related factor, including glare, blowing debris, animal or object 8 57.2% 0 0% 11 11.2% in roadway Total 14 100% 14 100% 98 100%456 baseline events, each 30 s long, was selected. Data reduc- tionists used the data directory and coded a variety of variables from these 456 randomly selected baseline driving events or brief driving periods. One random baseline event was selected for each driver-week of data collection. Baseline events were described using many of the same variables used to describe safety-critical events. The goal of identifying baseline events was to provide a comparison between normal driving and driving during critical events. For example, the proportion of time spent driving under various conditions and the propor- tion of time drivers performed various behaviors (e.g., eating, drinking, talking on citizens band [CB] radio or cell phone) were compared across different situations (1). Because only the subject vehicle was equipped with data collection units, only the behavior of the driver in that vehicle was coded and documented. As shown in Table 8.6, the most frequent critical factor for crashes was an object in the roadway, followed by driver-related factors associated with recognition errors, decision errors, and performance errors; each con- stituted 20% of the total cases. Not surprisingly, almost all (75%) the tire strikes involved some type of improper turn. The next two largest categories of contributing factor for crashes: tire strikes are driver performance error and driver decision error, respectively. For near crashes, the most frequent factor is driver-related recognition errors; more than 40% of near crashes were caused by inattention or distraction. Of these near crashes, almost one-quarter involved the subjectdriver not seeing the other vehicle during a lane change or merge. Countermeasures A close examination of the event causations reveals that a significant portion of the crashes or near crashes happened because of driver errors, such as inattention, distraction, or judgment errors. To prevent these events, the driver’s response should be altered, his or her attention should be improved, or his or her driving habits should be corrected. The most frequently suggested functional countermeasures relating to modifying driver behavior include increasing driver recognition of specific highway crash threats (improving driver recognition of forward threats), increasing driver attention, improving driver situation awareness, and defensive driving. The team examined the video data from Project 5, and the results are listed in Table 8.7. Most events are preventable by modifying driver behavior or increasing the attention level. The percentage indicates the portion of crashes and near crashes that would have been avoided if the suggested countermeasures had been applied. It was not unusual that more than one countermeasure could be selected for an event when the contributing factor was a combination of factors. Therefore, the total may be more than 100%. Because of the massive size of the data, the countermeasures identification for Project 6 was not as detailed as for projects

68Table 8.6. Contributing Factors for Crashes in Project 8 Critical Factor Crashes Crashes: Tire Strikes Near Crashes Critical reason not coded to this vehicle 0 0% 0 0% 16 26.2% Driver-related factor (critical nonperformance errors), including sleep; heart attack 0 0% 0 0% 4 6.6% or other physical impairment of the ability to act; drowsiness, fatigue, or other reduced alertness; other critical nonperformance Driver-related factor (recognition errors), including inattention; internal distraction; 1 20% 0 0% 27 44.3% external distraction; inadequate surveillance (e.g., failure to look); other or unknown recognition error Driver-related factor (decision errors), including too fast or too slow; misjudgment 1 20% 1 12.5% 6 9.8% of gap; following too closely or unable to respond to unexpected actions; false assumption of other road user’s actions; apparently intentional sign or signal violation; illegal U-turn or other illegal maneuver; failure to turn on headlamps; inadequate evasive action; aggressive driving; other or unknown decision error Driver-related factor (performance errors), including panic or freezing; over- 1 20% 1 12.5% 5 8.2% compensation; poor directional control; other or unknown performance error Environment-related factor, including sign missing; view obstruction by roadway 0 0% 6 75% 2 3.2% design; roadway geometry; sight distance; maintenance problems; slick roads; other highway-related conditions Environment-related factor, including: glare, blowing debris, animal or object 2 40% 0 0% 1 1.6% in roadway Total 5 100% 8 100% 61 100%discussed earlier. The team did differentiate avoidable crashes from unavoidable or likely avoidable crashes and near crashes, as shown in Table 8.8. Almost 40% of crashes can or are likely to be prevented. More than 80% of near crashes can or are likely to be prevented, given reasonable countermeasures. The detailed countermeasures to safety-critical events in Project 7 and Project 8 are illustrated in Tables 8.9 and 8.10 (1; 2), respectively. Because of multiple countermeasures appli- cable to one event, the total may be more than 100%. Tables 8.9 and 8.10 list the functional countermeasures that describe an intervention into the driving situation. Specifically, to technologically modify drivers’ behaviors, warning systems can be used to alert drivers so that they are not distracted or to correct driving habits to improve safety. Many of these systems, some of which are provided in the following list, have been tested in previous studies as technical countermeasures.which consists of a single analog black-and-white camera, a personal computer with a frame-grabber card, and an interface-to-vehicle network for obtaining ground speed. Distance from the center of the car to left and right lane markings, the angular offset between the car centerline and road centerline, the approximate road curvature, and marking characteristics can be measured and calculated to determine whether the car remains in the lane or is cross- ing lines. A similar system was developed by UMTRI. A lane-tracking system can be used mainly to alert a driver in circumstances of decreased alertness. It can also be used to correct drivers’ recognition errors, decision errors, and performance errors (1–3). 3. ACC System. Instead of simply maintaining a preset tar- get speed, as does a CCC system, an ACC system is prima- rily a speed and headway controller. It can modulate speed and use throttle and brakes simultaneously to manage headway to the leading vehicle. Usually an ACC system holds a maximum braking authority. In Project 2, this value is 0.3 g. When the headway decreases to the point at which the maximum braking response is required, the system will issue an alert or apply braking. This system will benefit drivers with recognition errors, a decreased alertness level, and some extent of decision errors and performance errors (3). 4. CSWS. A CSWS can usually help drivers slow down to a safe speed before entering an upcoming curve. It uses GPS and1. FCW System. An FCW system measures the headway between the subject vehicle and the leading vehicle. It issues a visual or audio warning when the equipped vehicle approaches the leading vehicle too rapidly. The system is effective in correcting drivers’ performance errors and decision errors. It is even more effective in alerting an inattentive or a less-alert driver (3). 2. Lane Tracking System/LDWS. A lane-tracking system or an LDWS usually can measure lane-keeping behavior. For example, VTTI developed a lane tracker called Road Scout,

69Table 8.7. Countermeasures by Category for Project 5 Crashes and Near Crashes Freeway Arterial Total No countermeasure applicable 5 6.5% 6 26.0% 11 11.0% 1. Increase driver alertness (reduce drowsiness) 2 2.6% 0 0.0% 2 2.0% 2. Prevent drift lane departures 1 1.3% 0 0.0% 1 1.0% 3. Improve vehicle control on curves 0 0.0% 0 0.0% 0 0.0% 4. Improve vehicle control on slippery road surfaces 0 0.0% 0 0.0% 0 0.0% 5. Improve vehicle control during braking 0 0.0% 0 0.0% 0 0.0% 6. Improve vehicle control during evasive steering 0 0.0% 0 0.0% 0 0.0% 7. Increase driver attention to forward scene 15 19.5% 1 4.3% 16 16.0% 8. Improve driver use of mirrors or provide better information from mirrors 0 0.0% 0 0.0% 0 0.0% 9. Improve general driver situation awareness and defensive driving 1 1.3% 0 0.0% 1 1.0% 10. Reduce travel speed 1 1.3% 0 0.0% 1 1.0% 11. Reduce speed on downgrades 0 0.0% 0 0.0% 0 0.0% 12. Reduce speed on curves or turns 0 0.0% 0 0.0% 0 0.0% 13. Reduce speed at or on exits (including ramps) 0 0.0% 0 0.0% 0 0.0% 14. Limit top speed to 70 mph (except on downgrades) 0 0.0% 0 0.0% 0 0.0% 15. Increase driver recognition of specific highway crash threats: stopped vehicle(s) in lane 14 18.2% 0 0.0% 14 14.0% ahead, traveling in same direction 16. Increase driver recognition of specific highway crash threats: moving or decelerating 55 71.4% 17 73.9% 72 72.0% vehicle(s) in lane ahead, traveling in same direction 17. Increase driver recognition of specific highway crash threats: vehicle in left adjacent lane 2 2.6% 0 0.0% 2 2.0% on highway 18. Increase driver recognition of specific highway crash threats: vehicle in right adjacent lane 1 1.3% 1 4.3% 2 2.0% on highway 19. Increase driver recognition of specific highway crash threats: vehicle in left adjacent lane 0 0.0% 0 0.0% 0 0.0% during merging maneuver 20. Increase driver recognition of specific highway crash threats: vehicle in right adjacent lane 0 0.0% 0 0.0% 0 0.0% during merging maneuver 21. Increase driver recognition or gap judgment recrossing or oncoming traffic at intersections 0 0.0% 0 0.0% 0 0.0% 22. Improve driver response execution of crossing or turning maneuver at intersections 0 0.0% 0 0.0% 0 0.0% (performance failure) 23. Improve driver recognition or gap judgment response execution at intersection 0 0.0% 0 0.0% 0 0.0% 24. Improve driver compliance with intersection traffic signal controls (both intentional and 0 0.0% 0 0.0% 0 0.0% unintentional intersection control violations) 25. Improve driver compliance with intersection traffic sign controls 0 0.0% 0 0.0% 0 0.0% 26. Increase forward headway during vehicle following 9 11.7% 2 8.7% 11 11.0% 27. Improve driver night vision in the forward field 0 0.0% 0 0.0% 0 0.0% 28. Provide warning to prevent rear encroachment or tailgating by other vehicle 0 0.0% 0 0.0% 0 0.0% 29. Provide advisory to driver regarding reduced road-tire friction (i.e., associated with 0 0.0% 0 0.0% 0 0.0% slippery roads) 30. Prevent vehicle mechanical failure 0 0.0% 0 0.0% 0 0.0% (continued on next page)

7031. Prevent splash and spray from this vehicle affecting other vehicle(s) 0 0.0% 0 0.0% 0 0.0% 32. Improve driver recognition or gap judgment relating to oncoming vehicle during 0 0.0% 0 0.0% 0 0.0% passing maneuver 33. Prevent animals from crossing roadways 0 0.0% 0 0.0% 0 0.0% 34. Provide driver with navigation system 0 0.0% 0 0.0% 0 0.0% 35. Aid to vertical clearance estimation 0 0.0% 0 0.0% 0 0.0% 36. Prevent or reduce trailer off-tracking outside travel lane or path 0 0.0% 0 0.0% 0 0.0% 97. Provide advance warning of need to stop at traffic sign or signal 0 0.0% 0 0.0% 0 0.0% 98. Driver error or vehicle failure apparent but countermeasure unknown 0 0.0% 0 0.0% 0 0.0% 99. Unknown 0 0.0% 0 0.0% 0 0.0% Total 101 131.2% 21 91.3% 133 133.0% Events Total 77 100% 23 100% 100 100.00% Table 8.7. Countermeasures by Category for Project 5 (continued) Crashes and Near Crashes Freeway Arterial TotalTable 8.8. Preventability of Crashes in Project 6 Avoidable Likely Avoidable Unavoidable Near Near Near Amendable Factor Crashes Crashes Crashes Crashes Crashes Crashes Correct driver-related factor, including sleep; drowsiness 4 (5.8%) 20 (2.6%) 0 (0.0%) 1 (0.1%) 4 (5.8%) 12 (1.6%) or other reduced alertness; other critical nonperformance Correct driver-related factor, including inattention; internal 13 (18.8%) 109 (14.3%) 0 (0.0%) 29 (3.8%) 9 (13.0%) 63 (8.3%) distraction; external distraction; inadequate surveillance (e.g., failure to look); other or unknown recognition error Correct driver-related factor, including too fast or too slow; 6 (8.7%) 134 (17.6%) 1 (1.4%) 60 (7.9%) 12 (17.4%) 24 (3.2%) misjudgment of gap; following too closely to respond to unexpected actions; false assumption of other road user’s actions; apparently intentional sign or signal violation; illegal U-turn or other illegal maneuver; failure to turn on headlamps; inadequate evasive action; aggressive driving; other or unknown decision error Correct driver-related factor, including poor directional 0 (0.0%) 5 (0.7%) 0 (0.0%) 20 (2.6%) 1 (1.4%) 5 (0.7%) control; other or unknown performance error Correct environment-related factor, including sign missing; 0 (0.0%) 12 (1.6%) 0 (0.0%) 14 (1.8%) 2 (2.9%) 0 (0.0%) view obstruction by roadway design; roadway geometry; sight distance; maintenance problems; slick roads; other highway-related conditions Correct environment-related factor, including glare, 0 (0.0%) 9 (1.2%) 0 (0.0%) 20 (2.6%) 4 (5.8%) 0 (0.0%) blowing debris, animal or object in roadway. Not correctable: crashes or near crashes caused by others 3 (4.3%) 33 (4.3%) 0 (0.0%) 164 (21.6%) 10 (14.5%) 27 (3.5%) Total 26 (38%) 322 (42%) 1 (1%) 308 (40%) 42 (61%) 131 (17%)

71Table 8.9. Countermeasures by Category for Project 7 Crashes: Crashes Tire Strikes Near Crashes No countermeasure applicable 1 7.1% 0 0.0% 18 18.4% 1. Increase driver alertness (reduce drowsiness) 3 21.4% 0 0.0% 7 7.1% 3. Prevent drift lane departures 0 0.0% 0 0.0% 4 4.1% 4. Improve vehicle control on curves 0 0.0% 0 0.0% 0 0.0% 5. Improve vehicle control on slippery road surfaces 0 0.0% 0 0.0% 0 0.0% 6. Improve vehicle control during braking 0 0.0% 0 0.0% 0 0.0% 7. Improve vehicle control during evasive steering 0 0.0% 0 0.0% 1 1.0% 8. Increase driver attention to forward scene 3 21.4% 0 0.0% 28 28.6% 9. Improve driver use of mirrors or provide better information from mirrors 0 0.0% 0 0.0% 3 3.1% 10. Improve general driver situation awareness and defensive driving 1 7.1% 0 0.0% 9 9.2% 12. Reduce travel speed 0 0.0% 0 0.0% 3 3.1% 13. Reduce speed on downgrades 0 0.0% 0 0.0% 0 0.0% 14. Reduce speed on curves or turns 0 0.0% 0 0.0% 1 1.0% 15. Reduce speed at or on exits (including ramps) 0 0.0% 0 0.0% 1 1.0% 16. Limit top speed to 70 mph (except on downgrades) 0 0.0% 0 0.0% 0 0.0% 17. Increase driver recognition of specific highway crash threats: stopped vehicle(s) in 1 7.1% 0 0.0% 4 4.1% lane ahead, traveling in same direction 18. Increase driver recognition of specific highway crash threats: moving or decelerating 0 0.0% 0 0.0% 6 6.1% vehicle(s) in lane ahead, traveling in same direction 19. Increase driver recognition of specific highway crash threats: vehicle in left adjacent 0 0.0% 0 0.0% 3 3.1% lane on highway 20. Increase driver recognition of specific highway crash threats: vehicle in right adjacent 0 0.0% 0 0.0% 1 1.0% lane on highway 21. Increase driver recognition of specific highway crash threats: vehicle in left adjacent 0 0.0% 0 0.0% 0 0.0% lane during merging maneuver 22. Increase driver recognition of specific highway crash threats: vehicle in right adjacent 0 0.0% 0 0.0% 0 0.0% lane during merging maneuver 23. Increase driver recognition or gap judgment regarding crossing or oncoming traffic 0 0.0% 0 0.0% 1 1.0% at intersections 25. Improve driver response execution of crossing or turning maneuver at intersections 2 14.3% 5 35.7% 2 2.0% (performance failure) 26. Improve driver recognition or gap judgment response execution at intersection 0 0.0% 0 0.0% 0 0.0% 27. Improve driver compliance with intersection traffic signal controls (both intentional 0 0.0% 0 0.0% 0 0.0% and unintentional intersection control violations) 28. Improve driver compliance with intersection traffic sign controls 0 0.0% 0 0.0% 0 0.0% 29. Increase forward headway during vehicle following 0 0.0% 0 0.0% 2 2.0% 30. Improve driver night vision in the forward field 0 0.0% 0 0.0% 5 5.1% 32. Provide warning to prevent rear encroachment or tailgating by other vehicle 0 0.0% 0 0.0% 0 0.0% 33. Provide advisory to driver regarding reduced road-tire friction (i.e., associated with 0 0.0% 0 0.0% 0 0.0% slippery roads) 34. Prevent vehicle mechanical failure 0 0.0% 0 0.0% 0 0.0% 36. Prevent splash and spray from this vehicle affecting other vehicle(s) 0 0.0% 0 0.0% 0 0.0% (continued on next page)

7237. Improve driver recognition or gap judgment relating to oncoming vehicle during 0 0.0% 0 0.0% 0 0.0% passing maneuver 38. Prevent animals from crossing roadways 4 28.6% 0 0.0% 9 9.2% 39. Provide driver with navigation system 2 14.3% 1 7.1% 1 1.0% 40. Aid to vertical clearance estimation 2 14.3% 0 0.0% 0 0.0% 41. Prevent or reduce trailer off-tracking outside travel lane or path 0 0.0% 7 50.0% 5 5.1% 42. Provide advance warning of need to stop at traffic sign or signal 0 0.0% 0 0.0% 3 3.1% 98. Driver error or vehicle failure apparent but countermeasure unknown 1 7.1% 2 14.3% 0 0.0% 99. Unknown 1 7.1% 0 0.0% 2 2.0% Total 21 150.0% 15 107.1% 119 121.4% Events Total 14 100% 14 100% 98 100% Table 8.9. Countermeasures by Category for Project 7 (continued) Crashes: Crashes Tire Strikes Near CrashesTable 8.10. Countermeasures by Category for Project 8 Crashes: Near Crashes Tire Strikes Crashes No countermeasure applicable 0 0% 0 0% 11 18% 1. Increase driver alertness (reduce drowsiness) 0 0% 1 13% 7 11% 2. Prevent drift lane departures 0 0% 0 0% 18 30% 3. Improve vehicle control on curves 0 0% 0 0% 1 2% 4. Improve vehicle control on slippery road surfaces 0 0% 0 0% 1 2% 5. Improve vehicle control during braking 0 0% 0 0% 0 0% 6. Improve vehicle control during evasive steering 0 0% 0 0% 0 0% 7. Increase driver attention to forward scene 0 0% 0 0% 16 26% 8. Improve driver use of mirrors or provide better information from mirrors 2 40% 0 0% 11 18% 9. Improve general driver situation awareness and defensive driving 3 60% 2 25% 23 38% 10. Reduce travel speed 0 0% 0 0% 2 3% 11. Reduce speed on downgrades 0 0% 0 0% 0 0% 12. Reduce speed on curves or turns 0 0% 0 0% 1 2% 13. Reduce speed at or on exits (including ramps) 0 0% 0 0% 0 0% 14. Limit top speed to 70 mph (except on downgrades) 0 0% 0 0% 0 0% 15. Increase driver recognition of specific highway crash threats: stopped vehicle(s) in lane ahead, 0 0% 0 0% 0 0% traveling in same direction 16. Increase driver recognition of specific highway crash threats: moving or decelerating vehicle(s) 0 0% 0 0% 3 5% in lane ahead, traveling in same direction 17. Increase driver recognition of specific highway crash threats: vehicle in left adjacent lane 0 0% 0 0% 9 15% on highway 18. Increase driver recognition of specific highway crash threats: vehicle in right adjacent lane 1 20% 0 0% 6 10% on highway 19. Increase driver recognition of specific highway crash threats: vehicle in left adjacent lane during 0 0% 0 0% 3 5% merging maneuver (continued on next page)

7320. Increase driver recognition of specific highway crash threats: vehicle in right adjacent lane 0 0% 0 0% 1 2% during merging maneuver 21. Increase driver recognition or gap judgment regarding crossing or oncoming traffic 0 0% 0 0% 0 0% at intersections 22. Improve driver response execution of crossing or turning maneuver at intersections 0 0% 1 13% 0 0% (performance failure) 23. Improve driver recognition or gap judgment response execution at intersection 0 0% 0 0% 0 0% 24. Improve driver compliance with intersection traffic signal controls (both intentional and unintentional intersection control violations) 0 0% 0 0% 1 2% 25. Improve driver compliance with intersection traffic sign controls 0 0% 0 0% 0 0% 26. Increase forward headway during vehicle following 0 0% 0 0% 2 3% 27. Improve driver night vision in the forward field 0 0% 0 0% 0 0% 28. Provide warning to prevent rear encroachment or tailgating by other vehicle 0 0% 0 0% 0 0% 29. Provide advisory to driver regarding reduced road-tire friction (i.e., associated with slippery roads) 0 0% 0 0% 0 0% 30. Prevent vehicle mechanical failure 0 0% 0 0% 0 0% 31. Prevent splash and spray from this vehicle affecting other vehicle(s) 0 0% 0 0% 0 0% 32. Improve driver recognition or gap judgment relating to oncoming vehicle during passing maneuver 0 0% 0 0% 0 0% 33. Prevent animals from crossing roadways 1 20% 0 0% 1 2% 34. Provide driver with navigation system 1 20% 1 13% 1 2% 35. Aid to vertical clearance estimation 0 0% 0 0% 0 0% 36. Prevent or reduce trailer off-tracking outside travel lane or path 0 0% 2 25% 0 0% 97. Provide advance warning of need to stop at traffic sign or signal 0 0% 5 63% 1 2% 98. Driver error or vehicle failure apparent but countermeasure unknown 0 0% 1 13% 0 0% 99. Unknown 0 0% 0 0% 0 0% Total 8 160% 13 163% 119 195% Event Total 5 100% 8 100% 61 100% Table 8.10. Countermeasures by Category for Project 8 (continued) Crashes: Near Crashes Tire Strikes Crashesdigital maps to anticipate curve locations and radiuses. Combined with recent driver control actions (turning signal and lateral acceleration), it determines if it is appropriate to issue a warning. The system is effective in correcting driver performance errors, recognition errors, and decision errors. It also helps to alert drivers of upcoming changes in road- way geometry (4). 5. Dilemma Zone Mitigation (DZM). Many crashes that occur at signalized intersections are associated with dilemma zones; for example, when faced with a yellow light, some drivers may decide to proceed through and others may decide to stop. Components of DZM usually include a carefully designed signal-timing cycle with an effective vehicle detection system that will identify the speed and size of vehicles, as well as the distance to the stopping line and provide additional safety by extending greentime to allow safe passage through the intersection if necessary (5). 6. Lateral Vehicle Detection (LVD). LVD usually consists of lateral cameras, a lane change assistance system, and a lateral collision warning system. The main purpose of LVD is to aid drivers to detect movements of vehicles in adjacent lanes and conduct corresponding maneuvers. The system will issue a warning when it determines that a lateral vehi- cle is trying to cut in front of the subject vehicle in an unsafe way (6). 7. Intelligent Speed Adaption System. In this system, developed in Sweden, GPS was used to locate a car on a digital map. The speed limit on that roadway was retrieved from the database, and the real speed of the vehicle was compared with the speed limit. The system adopts interventions that are preprogrammed in the vehicle (7).

74Besides these existing warning systems, potentially bene- ficial warning systems not yet tested might be effective in reducing safety-related events; for example, a system that is capable of detecting weather and road surface conditions (e.g., rainfall amount, snow amount, visibility, wet road surface) and proposing possible road friction parameter variations because of these conditions in order to issue corresponding warnings, and a customized warning system initiated by the user’s individual car key, which can adjust warning-issuing threshold values according to different driving habits. When making countermeasure recommendations, it should be recognized that emerging driver assistance systems may initiate some complexities and, therefore, the assessment of safety benefits is not straightforward. For example, when drivers rely on these safety systems, failure of such systems can be fatal. Incorporating some other countermeasures in a systematic approach will more than likely be beneficial. In conclusion, collision prevention should include a better design of roads, a more comprehensive recovery system, and a more coordinated safety management system. According to a report from the Organisation for Economic Co-operation and Development, some basic enforcement may be highly efficient. Seat belt usage, speed management, extra efforts to monitor high-risk drivers, and identification and monitoring of dangerous locations are all effective countermeasures that contribute to improvements in transportation system safety (8). Conclusions To determine the feasibility of using in-vehicle video data to make inferences about driver behavior that would allow investigation of the relationship between observable driver behavior and nonrecurring congestion to improve travel time reliability, the team explored the identified data sets to inves- tigate the usefulness of video and other supplementary data, proposed models for the estimation of travel time reliability measures, and identified potential problems in current data sources. Based on the analysis of the six naturalistic data sources, this study demonstrates the following: 1. It is feasible to identify driver behavior before near crashes and crashes from video data collected in a naturalistic driving study and thus infer the causes of those events. 2. Recommendations can be made to change driver behavior and, therefore, prevent (or reduce) crashes and near crashes. 3. Naturalistic data are useful to identify impacts of crashes on traffic conditions. Given the small sample of crashes and the fact that the DAS does not gather data when the engine is off, it is not possible to study the impact of inci- dents on travel time reliability using in-vehicle data alone. When effectively integrated with external data sources, however, which is extremely feasible, given an accuratetime and location stamp in the data set, naturalistic data can be highly efficient in recognizing the relationship between modifying driver behavior and nonrecurring congestion. 4. Increased coordination with weather and traffic volume data is required to determine when nonrecurring conges- tion exists and which driver actions result from these non- recurring events. 5. It is possible to analyze naturalistic driving data to char- acterize typical levels of variability in travel times and develop measures for quantifying travel time reliability. Although the team has successfully proved the feasibility of using video and other in-vehicle and external data to study driver behavior and related nonrecurring congestion, some limitations need to be enhanced when a full data analysis is conducted. These limitations are summarized as follows: 1. A limited number of safety-related events exist in the data sets the team examined because of the naturalistic nature of the data. This shortcoming can be improved by extend- ing the time duration of data collection or increasing the number of participants. Both can be realized in the SHRP 2 Safety Project S07 study, in which a much larger data col- lection effort will be performed. 2. The external data sources in this study were examined only for availability and accuracy. Because of time constraints, no real connection was conducted to relate driver behavior to external driving environment. 3. Because of the limited size of travel time data in the natu- ralistic data sets, other data sets were also used to develop travel time reliability models. These models are general and apply regardless of the source of travel time data. These limitations can be corrected if a larger video data set can be collected or the external data can be better linked with the in-vehicle data, which will be feasible in the next stage of the study. Recommendations and Discussion An important component of the next stage of SHRP 2 research is a large-scale naturalistic driving data collection project (Project S07). The field study data collection contractors will be responsible for advertising for participants with preprepared recruitment materials, scheduling participant drivers for installation and assessment, conducting driver intake testing and installing the DAS in the owner’s vehicle, collecting data, addressing problems encountered during the study, investi- gating crashes, transmitting data, carrying out quality control procedures, and preparing periodic reports that document field study activities. The combined goal is to collect approx-

75imately 4,000 vehicle-years of data in a 30-month period. The following are the planned variables to be collected: 1. Antilock Brake System (ABS) Activation: Antilock brake activation indicator. 2. Acceleration, x axis: Vehicle acceleration in the longitudi- nal direction versus time. 3. Acceleration, x axis fast: Vehicle acceleration in the lon- gitudinal direction versus time. Fast buffer (−9 s to +3 s) based on trigger (e.g., in crash or other high-acceleration event). 4. Acceleration, y axis: Vehicle acceleration in the lateral direction versus time. 5. Acceleration, y axis fast: Vehicle acceleration in the lateral direction versus time. Fast buffer (−9 s to +3 s) based on trigger (e.g., in crash or other high-acceleration event). 6. Acceleration, z axis: Vehicle acceleration vertically (up or down) versus time. 7. Acceleration, z axis fast: Vehicle acceleration vertically (up or down) versus time. Fast buffer (−9 s to +3 s) based on trigger (e.g., in crash or other high-acceleration event); 8. Airbag, Driver: Indicates deployment of the driver’s airbag. 9. Alcohol: Presence of alcohol in the vehicle cabin. 10. Altitude, GPS: Altitude. 11. Audio: Audio recording for 30 s when incident button is pushed. 12. Average Fuel Economy after Fueling: Average fuel econ- omy after fueling. 13. Cruise Control: Status of cruise control. 14. Date: UTC year, month, and day. 15. Distance: Distance of vehicle travel. 16. Driver Button Flag: Flag indicating that the driver has pressed the incident button. 17. Electronic Stability Control (ESC): ESC activation indicator. 18. Engine RPM: Instantaneous engine speed. 19. Face, Driver ID: Machine-vision–based identification of the driver within those observed in a specific vehicle. The system observes within a vehicle to identify drivers who drive that vehicle (i.e., not a unique identification across all drivers in the study). 20. Face, Gaze Zone: Estimation of the location of the driver’s gaze categorized into zones in and around the vehicle. 21. Face, Gaze Zone Confidence: Confidence in the estimation of the zone at which the driver is looking. 22. Fuel Economy, Instantaneous: Instantaneous fuel economy. 23. Fuel Level: Fuel level. 24. Heading, GPS: Compass heading of vehicle from GPS. 25. Headlight Setting: State of headlamps. 26. Horn Status: Actuation of horn.27. Illuminance, Ambient: Ambient exterior light. 28. LDWS: Status of original equipment manufacturer (OEM) lane departure warning system. 29. Lane Marking, Distance, Left: Distance from vehicle centerline to inside of left side lane marker based on vehicle-based machine vision. 30. Lane Marking, Distance, Right: Distance from vehicle centerline to inside of right side lane marker based on vehicle-based machine vision. 31. Lane Marking, Probability, Right: Probability that vehicle- based machine vision lane marking evaluation is providing correct data for the right side lane markings. 32. Lane Marking, Type, Left: Type of lane marking imme- diately to the left of vehicle using vehicle-based machine vision. 33. Lane Marking, Type, Right: Type of lane marking imme- diately to the right of vehicle using vehicle-based machine vision. 34. Lane Marking, Probability, Left: Probability that vehicle- based machine vision lane marking evaluation is providing correct data for the left side lane markings. 35. Lane Position Offset: Distance to the left or right of the center of the lane based on machine vision. 36. Lane Width: Distance between the inside edge of the innermost lane marking and the left and right of the vehicle. 37. Latitude: Vehicle position latitude. 38. Longitude: Vehicle position longitude. 39. Pedal, Accelerator Position: Position of the accelerator pedal collected from the vehicle network and normalized using manufacturer specifications. 40. Pedal, Brake: On or off press of brake pedal. 41. Pitch Rate, y axis: Vehicle angular velocity around the lateral axis. 42. Pitch Rate, y axis fast: Vehicle angular velocity around the lateral axis. Fast buffer (−9 s to +3 s) based on trigger (e.g., in crash or other high-acceleration event). 43. P-R-N-D-L: Gear position. 44. Radar, Azimuth Forward: Angular measure to target. 45. Radar, Range Rate Forward: Range rate to forward radar targets. 46. Radar, Range, Forward: Range to forward radar targets measured from the radar to the targets. 47. Radar, Target Identification: Numerical value used to differentiate one radar target from others. 48. Radius of Curvature, Machine Vision: Estimation of road- way curvature based on machine vision. 49. Roll Rate, x axis: Vehicle angular velocity around the longitudinal axis. 50. Roll Rate, x axis fast: Vehicle angular velocity around the longitudinal axis. Fast buffer (−9 s to +3 s) based on trigger (e.g., in crash or other high-acceleration event).

7651. Satellites, Number of: Count of the number of satellites being used for GPS position fix. 52. Seat belt, Driver: Use of the seat belt by the driver. 53. Speed, GPS: Vehicle speed from GPS. 54. Speed, Vehicle Network: Vehicle speed indicated on speedometer collected from network. 55. Steering Wheel Position: Angular position and direction of the steering wheel from neutral position. 56. Sync: Integer used to identify one time sample of data when presenting rectangular data. 57. Temperature, Interior: Vehicle interior temperature. 58. Time: UTC Time. Local time offsets need to be applied. 59. Track Type: Classification of target based on radar. 60. Traction Control: Status of traction control system. 61. Turn Signal: State of illumination of turn signals. 62. Vehicle Angle Relative to Roadway: Vehicle angle relative to the roadway based on machine vision. 63. Video Frame: Frame number of video at point in time. 64. Video, Driver and Left Side View: Video capture of the driver and exterior area to the left of the vehicle. 65. Video, Forward Roadway: Video capture of forward roadway. 66. Video, Occupancy Snapshot: Occupancy snapshot. 67. Video, Rear View: Video capture to the rear of the vehicle. 68. Video, Right Side View: Video capture to the right of the vehicle. 69. Wiper Setting: Indicates setting of windshield wipers. 70. Yaw Rate, z axis: Vehicle angular velocity around the ver- tical axis. 71. Yaw Rate, z axis fast: Vehicle angular velocity around the vertical axis. Fast buffer (−9 s to +3 s) based on trigger (e.g., in crash or other high-acceleration event). To ensure that data collected in the SHRP 2 Safety Proj- ect S07 study are versatile and comprehensive enough to be used to conduct full-scaled research to study nonrecurring congestion related to driver behavior, several recommendations have resulted from the findings of this study. First, the procedure to recruit participants needs to be carefully designed. Ideally, a comprehensive population of drivers ranging evenly across every age category, income category, and occupation category should be included. When recruiting participants, it is crucial to make it clear to them that driver information is vital for the research. To better identify drivers, two methods can be used: 1. A formal statement needs to be included in the contract to make the signer the exclusive driver of the vehicle. 2. A touch-screen device can be installed on board to collect information before and after each trip. The touch-screen equipment can be designed so that a customized interface will be displayed to the driver to input trip-related infor-mation by selecting certain check boxes. The before-trip information-collecting interface may consist of a list of the first names of household members for the driver to select from, a list of trip purposes, weather conditions when the trip started, and any information about why the driver selected the time of departure. The after-trip information- collecting interface may include an “original trip purpose changed” option, a “route choice changed” option, and a “crash happened en route” option. Necessary hardware can be designed to connect the input touch-screen with the engine so that the driver can start the engine only after the information is input. To ensure safety while driving, the device should be disabled while the vehicle is in motion to prevent driver distraction. One major concern this type of device may impose on such studies is that it will remind drivers that they are being monitored and thus may reduce the naturalistic nature of the studies. Second, to serve the research purpose, certain data are more important than others. The following four categories are imperative: 1. Basic onboard equipment should include devices that col- lect the following data: video; vehicle network information (speed, brake pedal, throttle, and turn signal); GPS data (latitude, longitude, and heading); X, Y, and Z acceleration; distances between the subject and surrounding objects; lane location information (X, Y, and Z); driver behavior (seat belt usage, lights on or off); and yaw rate. 2. Video cameras should shoot at least five views: front, back, right, left, and the driver. The resolution of the video camera should be high enough to identify ongoing traffic conditions, weather conditions, and the driver’s hand movements and facial expressions. Correction of sun glare to improve video quality is available when needed. 3. The frequency setting should be high enough so that the video is continuous, the acceleration or deceleration of the vehicles should be clearly recorded, and the reaction times need to be recorded and measured. The recommended minimum frequency for GPS devices is 1 Hz and, for all other equipment, 10 Hz. 4. To improve the versatility of the data so that it can be used in other, related research, the vehicle performance parameters (e.g., engine speed, throttle position, and torque) should be recorded. Table 8.11 shows a sublist of variables that are vital to the next stage of this research and that will be collected in the Project S07 study. Units and minimum rates of data collection are suggested.Third, the data collection system needs to run for an addi- tional 10 min after the engine is turned off in case the vehicle is involved in an accident. During the additional data reduction,

77Table 8.11. Recommended Variables for Collection Recommended Variable Name Units Minimum Rate 1. Acceleration, x axis g 10 Hz 2. Acceleration, y axis g 10 Hz 3. Acceleration, z axis g 10 Hz 4. Altitude, GPS ft 1 Hz 5. Date NA NA 6. Distance mi NA 7. Engine RPM rpm NA 8. Face, Driver ID NA 10 Hz 9. Face, Gaze Zone NA 10 Hz 10. Fuel Economy, Instantaneous mpg NA 11. Heading, GPS degree 1 Hz 12. LDWS NA NA 13. Lane Marking, Distance, Left ft 10 Hz 14. Lane Marking, Distance, Right ft 10 Hz 15. Lane Marking, Type, Left NA 10 Hz 16. Lane Marking, Type, Right NA 10 Hz 17. Lane Position Offset ft NA 18. Lane Width ft NA 19. Latitude Ddd.sss 1 Hz 20. Longitude Ddd.sss 1 Hz 21. Pedal, Accelerator Position NA NA 22. Pedal, Brake NA NA 23. Pitch Rate, y axis degree/s 10 Hz 24. Radar, Azimuth Forward rad 10 Hz 25. Radar, Range, Forward ft 10 Hz 26. Radar, Target Identification NA 10 Hz 27. Radius of Curvature, NA NA Machine Vision 28. Roll Rate, x axis degree/s 10 Hz 29. Seat belt, Driver NA 10 Hz 30. Speed, GPS mph 1 Hz 31. Time NA NA 32. Track Type NA NA 33. Video Frame NA NA 34. Video, Driver and Left Side View NA 10 Hz 35. Video, Forward Roadway NA 10 Hz 36. Video, Occupancy Snapshot NA NA 37. Video, Rear View NA 10 Hz 38. Video, Right Side View NA NA 39. Wiper Setting NA 1 Hz 40. Yaw Rate, z axis degree/s 10 Hzdata collection was usually found to stop the instant the driver stopped the vehicle. It is important, however, to observe the traffic conditions being affected by a safety-related event. In discussion with the SHRP 2 S06 contractor, a potential safety hazard was identified that may deem this recommendation infeasible. Specifically, continued data collection after an accident may result in a vehicle explosion if the vehicle gasoline tank is jeopardized. Fourth, to improve the linking of vehicle data with exter- nal data, it is ideal to standardize the time and location data. For external data, the database in some states is built on the milepost system. The conversion of milepost locations to standard latitude and longitude coordinates should be con- ducted ahead of time. For vehicle data, the synchronized GPS clock should be used instead of the local computer time for better connection of the data with external traffic, crash, work zone, and weather data. Fifth, because a limited number of crashes occurred in all the candidate data sets—especially severe crashes that affected traffic conditions—certain adjustments are needed to create a statistically significant database. A lengthier data collection effort or more drivers involved in the study would be ideal. For example, the 2,500-Car Study (SHRP 2 Safety Project S07), which will soon be conducted, is a quality candidate. Another solution is simulation, which can be used to compensate for data shortage. Sixth, additional analysis of existing data is required to study typical levels of variability in driver departure times, typical levels of variability in trip travel times, and the level of variability in driver route choices. A characterization of this behavior is critical in attempting to quantify and develop travel time reliability measures because it identifies potential causes for travel time variability and thus can enhance travel time reliability models. These data may be augmented with tests on a driving simulator to study the impact of travel time reliability on driver route choice behavior. Finally, although a number of studies have used video cam- eras to gather data, an ideal starting point is a compiled data source list that summarizes existing video-involved studies with specifications of data collected, limitations of data usage, and access issues. Such a list would help prevent redundancy in future investigation efforts. This research can benefit from the data being collected under the IntelliDrive Program (IntelliDrive is a service mark of the U.S. Department of Transportation). The IntelliDrive Program is, as introduced on its website, “a multimodal initiative that aims to enable safe, interoperable networked wireless communications among vehicles, the infrastructure, and passenger’s personal communications devices” (9). It will collect and disseminate data, including roadway, traffic condi- tion, weather, crashes, and traffic control among vehicles. With the development of IntelliDrive, it is possible to use the data

78sets collected by the program to complement the scantiness of regular state-maintained traffic count and crash data. References 1. Lerner, N., J. Jenness, J. Singer, S. G. Klauer, S. Lee, M. Donath, M. Manser, and M. Ward. An Exploration of Vehicle-Based Monitoring of Novice Teen Drivers: Draft Report. Virginia Tech Transportation Institute, Blacksburg, Va., 2008. 2. Hickman, J. S., R. R. Knipling, R. L. Olson, M. C. Fumero, M. Blanco, and R. J. Hanowski. Phase I—Preliminary Analysis of Data Col- lected in the Drowsy Driver Warning System Field Operational Test: Task 5, Preliminary Analysis of Drowsy Driver Warning System Field Operational Test Data. NHTSA, 2005. 3. University of Michigan Transportation Research Institute. Auto- motive Collision Avoidance System Field Operational Test Report: Methodology and Results. Report DOT HS 809 900 NHTSA, 2005. 4. University of Michigan Transportation Research Institute. Road Departure Crash Warning System Field Operational Test: Methodology and Results. NHTSA, 2006.5. Dingus, T. SHRP 2 S05 Status Update and Current Design Plans. Pre- sented at SHRP2 Safety Research Symposium, Washington, D.C., 2008. http://onlinepubs.trb.org/onlinepubs/shrp2/TomDingus SymposiumPresentation.pdf. Accessed May 17, 2011. 6. Blanco, M., J. S. Hickman, R. L. Olson, J. L. Bocanegra, R. J. Hanowski, A. Nakata, M. Greening, P. Madison, G. T. Holbrook, and D. Bowman. Investigating Critical Incidents, Driver Restart Period, Sleep Quantity, and Crash Countermeasures in Commercial Vehicle Opera- tions Using Naturalistic Data Collection. FMCSA, 2008. 7. Fitch, G., H. Rakha, M. Arafeh, M. Blanco, S. Gupta, R. Zimmerman, and R. Hanowski. Safety Benefit Evaluation of a Forward Collision Warning System: Final Report. Report DOT HS 810 910. NHTSA, 2008. 8. International Transport Forum. Towards Zero: Ambitious Road Safety Targets and the Safe System Approach. Organisation for Economic Co-operation and Development, 2008. www.internationaltransport forum.org/Pub/pdf/08TowardsZeroE.pdf. Accessed May 17, 2011. 9. Research and Innovative Technology Administration, U.S. Depart- ment of Transportation. About Intelligent Transportation Systems. www.its.dot.gov/its_program/about_its.htm. Accessed May 17, 2011.

Next: Appendix A - Project 2 Data Dictionary »
Feasibility of Using In-Vehicle Video Data to Explore How to Modify Driver Behavior That Causes Nonrecurring Congestion Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

TRB’s second Strategic Highway Research Program (SHRP 2) Report S2-L10-RR-1: Feasibility of Using In-Vehicle Video Data to Explore How to Modify Driver Behavior That Causes Nonrecurring Congestion presents findings on the feasibility of using existing in-vehicle data sets, collected in naturalistic driving settings, to make inferences about the relationship between observed driver behavior and nonrecurring congestion.

The report, a product of the SHRP 2 Reliability focus area, includes guidance on the protocols and procedures for conducting video data reduction analysis.

In addition, the report includes technical guidance on the features, technologies, and complementary data sets that researchers can consider when designing future instrumented in-vehicle data collection studies.

The report also highlights a new modeling approach for travel time reliability performance measurement across a variety of traffic congestion conditions.

An e-book version of this report is available for purchase at Google, Amazon, and iTunes.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!