Learning Risk-aware Costmaps for Traversability in Challenging Environments

Abstract

One of the main challenges in autonomous robotic exploration and navigation in unknown and unstructured environments is determining where the robot can or cannot safely move. A significant source of difficulty in this determination arises from stochasticity and uncertainty, coming from localization error, sensor sparsity and noise, difficult-to-model robot-ground interactions, and disturbances to the motion of the vehicle. Classical approaches to this problem rely on geometric analysis of the surrounding terrain, which can be prone to modeling errors and can be computationally expensive. Moreover, modeling the distribution of uncertain traversability costs is a difficult task, compounded by the various error sources mentioned above. In this work, we take a principled learning approach to this problem. We introduce a neural network architecture for robustly learning the distribution of traversability costs. Because we are motivated by preserving the life of the robot, we tackle this learning problem from the perspective of learning tail-risks, i.e. the Conditional Value-at-Risk (CVaR). We show that this approach reliably learns the expected tail risk given a desired probability risk threshold between 0 and 1, producing a traversability costmap which is more robust to outliers, more accurately captures tail risks, and is more computationally efficient, when compared against baselines. We validate our method on data collected a legged robot navigating challenging, unstructured environments including an abandoned subway, limestone caves, and lava tube caves.

Published in RA-L with ICRA presentation option (IEEE International Conference on Robotics and Automation, 2022)

References in corpus (6)

• Stochastic Gradient Descent for Non-smooth Optimization: Convergence Results and Optimal Averaging Schemes
• Distributional Reinforcement Learning with Quantile Regression
• NeBula: Quest for Robotic Autonomy in Challenging Environments; TEAM CoSTAR at the DARPA Subterranean Challenge
• PLGRIM: Hierarchical Value Learning for Large-scale Exploration in Unknown Environments
• Distributionally robust risk map for learning-based motion planning and control: A semidefinite programming approach
• CVaR-based Flight Energy Risk Assessment for Multirotor UAVs using a Deep Energy Model

Cited by in corpus (1)

• Hybrid Imitative Planning with Geometric and Predictive Costs in Off-road Environments