ArtPlanner: Robust Legged Robot Navigation in the Field

Abstract

Due to the highly complex environment present during the DARPA Subterranean Challenge, all six funded teams relied on legged robots as part of their robotic team. Their unique locomotion skills of being able to step over obstacles require special considerations for navigation planning. In this work, we present and examine ArtPlanner, the navigation planner used by team CERBERUS during the Finals. It is based on a sampling-based method that determines valid poses with a reachability abstraction and uses learned foothold scores to restrict areas considered safe for stepping. The resulting planning graph is assigned learned motion costs by a neural network trained in simulation to minimize traversal time and limit the risk of failure. Our method 1 achieves real-time performance with a bounded computation time. We present extensive experimental results gathered during the Finals event of the DARPA Subterranean Challenge, where this method contributed to team CERBERUS winning the competition. It powered navigation of four ANYmal quadrupeds for 90 minutes of autonomous operation without a single planning or locomotion failure.

Learning robust autonomous navigation and locomotion for wheeled-legged robots

“…This understanding is crucial for issuing commands that optimize efficiency on flat terrain while maintaining agility when faced with obstacles. Many existing approaches (20,21) are based on explicit navigation costs, such as traversability (21,22), without considering the robot's whole-body states. They focus on generating kinematic navigation plans by sampling-based planning on these estimated cost maps.…”

“…In addition to the previous points, the higher speed capabilities of wheeled-legged robots introduce the need for shorter reaction times, which raises safety concerns and calls for more responsive control systems. State-of-the-art sampling-based planners designed for legged robots typically take several seconds to compute a path (20). However, when operating at speeds of multiple meters per second, relying on such planning methods would necessitate long foresight and could result in collisions in dynamic environments.…”

“…Such discontinuous or oscillatory behavior can compromise efficiency and hinder the robot's ability to respond to complex dynamic scenarios. In some cases, navigation failure may occur when computed navigation paths are not accurately followed (20).…”