The ATHLETE Footfall Planner is an automated analysis and user interface software to facilitate Tri-ATHLETE walking on extreme terrain. Footfall assists ground controllers by integrating situational awareness, collision free leg-motion planning, compliance modeling, sag-aware plan compensation, stability analysis, and gait generation. Footfall uses the Ames Stereo Pipeline (ARC) to processes images from multiple body-mounted cameras to construct digital elevation models (DEM) of the local terrain. DEMs are then analyzed for traversability, reachability, and stability (polygon of support and tip-over energy) using robot telemetry. Compliance from the robots joints and wheels is modeled and used to update the reported pose of the robot. An interactive 3D display, Verve (ARC), integrated into the ATHLETE Workbench, is used to visualize and select footfall locations. A bi-directional rapidly exploring random tree motion planner is then used to plan an optimized trajectory to the chosen location, while safely avoiding self and environmental collisions and kinematic constraints. Further technical details can be found in the papers below, with more recent results expected to be published during 2011.
The Footfall Planner has been used to control two iterations of the physical robot and has been tested at two analog field trials:
For more information about the robots, please see the ATHLETE website.
This paper from the Journal of Field Robotics gives the best overview of the Footfall Planner and summarizes the full multi-year effort.
Development and Field Testing of the FootFall Planning System for the ATHLETE Robots
Vytas SunSpiral, DW Wheeler, Daniel Chavez-Clemente, David Mittman
The Journal of Field Robotics, Vol. 29: pp 483–505. 2012.
This paper is an earlier summary paper:
FootFall: A Ground Based Operations Toolset Enabling Walking for the ATHLETE Rover
Vytas SunSprial, Daniel Chavez-Clemente, Michael Broxton, Leslie Keely, Patrick Milhelich
In proceedings, AIAA Space2008, San Diego, California, Sept. 2008.
The rest of these papers focus on detailed aspects of the Footfall Planner:
“FootSpring: A Compliance Model for the ATHLETE Family of Robots”
Dawn Wheeler, Daniel Chavez-Clemente, Vytas SunSpiral,
In proceedings, 10th International Symposium on Artificial Intelligence, Robotics and Automation in Space, (i-SAIRAS 2010). Sapporo, Japan, August 2010.
“A Practical Comparison of Motion Planning Techniques for Robotic Legs in Environments with Obstacles”
T. Smith, D. Chavez-Clemente,
In Proceedings of the Third IEEE International Conference on Space Mission Challenges for Information Technology. Pasadena, CA. July 2009.
Athlete's Feet: Multi-Resolution Planning for a Hexapod Robot
Tristan Smith, Javier Barreiro, David Smith, Vytas Sunspiral, Daniel Chavez,
In proceedings, International Conference on Automated Planning and Scheduling (ICAPS), Sydney, Australia, Sept. 2008.