Learning from Demonstration for Autonomous Navigation in Complex Unstructured Terrain
International Journal of Robotics Research
Path planning with variable-fidelity terrain assessment
Robotics and Autonomous Systems
Autonomous over-the-horizon navigation using LIDAR data
Autonomous Robots
Terrain traversability analysis methods for unmanned ground vehicles: A survey
Engineering Applications of Artificial Intelligence
| Hi-index | 0.01 |
In January 2004, NASA's twin Mars Exploration Rovers (MERs), Spirit and Opportunity, began searching the surface of Mars for evidence of past water activity. To localize and approach scientifically interesting targets, the rovers employ an onboard navigation system. Given the latency in sending commands from Earth to the Martian rovers (and in receiving return data), a high level of navigational autonomy is desirable. Autonomous navigation with hazard avoidance (AutoNav) is currently performed using a local path planner called GESTALT (grid-based estimation of surface traversability applied to local terrain) that incorporates terrain and obstacle information generated from stereo cameras. GESTALT works well at guiding the rovers around narrow and isolated hazards; however, it is susceptible to failure when clusters of closely spaced, nontraversable rocks form extended obstacles. In May 2005, a new technology task was initiated at the Jet Propulsion Laboratory to address this limitation. Specifically, a version of the Field D* global path planner was integrated into MER flight software, enabling simultaneous local and global planning during AutoNav. A revised version of AutoNav was then uploaded to the rovers during the summer of 2006. In this paper we describe how this integration of global planning into the MER flight software was performed and provide results from both the MER surface system test bed rover and five fully autonomous runs by Opportunity on Mars. © 2009 Wiley Periodicals, Inc.