provide images from an elevated perspective, plus a fixed frame of reference) will encounter considerable problems if they do not have rapidly produced maps and flexible navigation tools. Capable imaging systems (both descent and rover-mounted) are necessary, as are improvements in autonomous navigation systems and communications. Scientific goals that require ranging over tens or hundreds of kilometers may be difficult to accomplish except in cases of relatively unchallenging geologic terrains. These views are echoed by the Mars Surveyor 2001 Science Definition Team (SDT), whose report comments that the lack of a satisfactory robotic field capability, even when the vehicle is under the full control of mission scientists, is "somewhat sobering."57 The SDT ascribes the limitations not to the rover hardware, but to the following:
It is clear that field demonstrations are essential. One area of particular importance is the operational integration of autonomous and direct control systems. A sophisticated rover should, in principle, be able to make observations while traversing from one predetermined location to another and use the results of the observations to select the most scientifically interesting route to take. Recent field tests have demonstrated that this capacity does not yet exist even when scientists have full control of the rover.58 Limited communications windows and available bandwidth will, necessarily, limit the degree to which mission scientists can have direct control over the planning of the rover's operations. Without some degree of autonomous control during the periods when it is not in communication with Earth, the rover is likely to spend a significant fraction of its operational life waiting for instructions. With the twice-daily communications sessions scheduled for future Mars Surveyor missions, this downtime could amount to 90 percent. Field demonstrations offer a ready means to develop and validate schemes for autonomous operations and to develop techniques for their harmonious integration with the limited periods when mission scientists will be in the control loop.
What also is clear is that the usefulness of field demonstrations will be greatly enhanced if better continuity exists between tests so that problems exposed in one test are not "forgotten" during the planning of subsequent tests. Many of the problems of this type that arose in the various rover tests conducted to date were related to people and operational systems rather than to the rovers themselves. These problems are, therefore, likely to be universal with respect to mobile platforms or mission objectives. Operational problems experienced by the Sojourner team (e.g., the fact that it took them much longer than planned to maneuver the rover from one rock to the next) could have been anticipated and planned for had there been better communication of field-test results. A mechanism needs to be devised to ensure that important operational problems are known and acted on by all groups. Adequate peer review of proposed operational tests and the prompt publication of the results of those tests in peer-reviewed journals are essential.
1. Space Studies Board, National Research Council, An Integrated Strategy for the Planetary Sciences: 1995–2010, National Academy Press, Washington, D.C., 1994.
2. Roadmap Development Team, National Aeronautics and Space Administration, Mission to the Solar System: Exploration and Discovery—A Mission and Technology Roadmap, Version B, Jet Propulsion Laboratory, Pasadena, Calif., 1996.
3. Extensive compilations of references to developments in rover technology can be found on the home pages of the following organizations: the Robotics Institute <http://www.ri.cmu.edu> at Carnegie Mellon University, the Artificial Intelligence Laboratory <http://www.ai.mit.edu> and the Field and Space Robotics Laboratory <http://robots.mit.edu> at the Massachusetts Institute of Technology, the Mobile Robotics Department <http://www.sandia.gov/Robotic_Range> at Sandia National Laboratories, and the Rover and Telerobotics Program <http://robotics.jpl.nasa.gov> at the Jet Propulsion Laboratory.
4. D.S. Pivirotto and W.C. Dias, United States Planetary Rover Status, JPL Publication 90-6, Jet Propulsion Laboratory, Pasadena, Calif., 1989.