Legged robot navigation in unstructured terrain
Developing footstep planning methods to produce foothold placements using visual perception and proper environment modelling.
12 December 2019
Legged robots have gained a lot of locomotion capabilities in the past few years, especially in the control level. Navigation over complex and unstructured environments using exteroceptive perception is still an active research topic. We have developed footstep planning methods to produce foothold placements, using visual perception and proper environment modelling, given a black box walking controller.
We achieved this by integrating scene segmentation for rough terrain surfaces and path planning analysis based on range input data. The system is experimentally validated using real-world legged robots for challenging terrains, such as the UCL RPBP, IIT COMAN, IIT WALK-MAN, IIT CENTAURO, and IIT COMAN+ humanoid and animaloid robots.
Relevant publications
- D. Kanoulas, N. Tsagarakis, and M. Vona "Curved Patch Mapping and Tracking for Irregular Terrain Modeling: Application to Bipedal Robot Foot Placement" Elsevier Robotics and Autonomous Systems, RAS 2019.
- V. Suryamurthy, V. Raghavan, A. Laurenzi, N. Tsagarakis, and D. Kanoulas, "Terrain Segmentation and Roughness Estimation using RGB Data: Path Planning Application on the CENTAURO Robot" In the 19th IEEE/RAS International Conference on Humanoid Robots, Humanoids 2019.
- V. Raghavan, D. Kanoulas, A. Laurenzi, D. Caldwell, and N. Tsagarakis, "Variable Configuration Planner for Legged-Rolling Obstacle Negotiation Locomotion: Application on the CENTAURO Robot" In the 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2019.
- D. Kanoulas, A. Stumpf, V. Raghavan, C. Zhou, A. Toumpa, O. von Stryk, D. Caldwell, and N. Tsagarakis, "Footstep Planning in Rough Terrain for Bipedal Robots using Curved Contact Patches" In the 2018 IEEE International Conference on Robotics and Automation, ICRA 2018.
- D. Kanoulas C. Zhou, A. Nguyen, G. Kanoulas, D. Caldwell, and N. Tsagarakis, "Vision-Based Foothold Contact Reasoning using Curved Surface Patches" In the 17th IEEE/RAS International Conference on Humanoid Robots, Humanoids 2017. Best Interactive Paper Award Winner.