Autonomous Off-road Navigation over Extreme Terrains with Perceptually-challenging Conditions

Abstract

We propose a framework for resilient autonomous navigation in perceptually challenging unknown environments with mobility-stressing elements such as uneven surfaces with rocks and boulders, steep slopes, negative obstacles like cliffs and holes, and narrow passages. Environments are GPS-denied and perceptuallydegraded with variable lighting from dark to lit and obscurants (dust, fog, smoke). Lack of prior maps and degraded communication eliminates the possibility of prior or off-board computation or operator intervention. This necessitates real-time onboard computation using noisy sensor data. To address these challenges, we propose a resilient architecture that exploits redundancy and heterogeneity in sensing modalities. Further resilience is achieved by triggering recovery behaviors upon failure. We propose a fast settling algorithm to generate robust multi-fidelity traversability estimates in real-time. The proposed approach was deployed on multiple physical systems including skid-steer and tracked robots, a high-speed RC car and legged robots, as a part of Team CoSTAR's effort to the DARPA Subterranean Challenge, where the team won 2 nd and 1 st place in the Tunnel and Urban Circuits, respectively.

Blending Efficiency and Resilience in the Performance Assessment of Urban Intersections: A Novel Heuristic Informed by Literature Review

“…Incorporating Quantum Key Distribution into vehicle-to-infrastructure networks safeguards against quantum computational advances, ensuring secure data communication in urban settings [44]. A similar concern is tackled by [45], focusing on a novel framework for robust autonomous navigation in challenging, unfamiliar environments with mobility-stressing factors like uneven terrain, steep slopes, negative obstacles, and limited visibility due to dust or fog. The system addressed the absence of GPS, degraded perception, and the lack of prior maps, necessitating real-time on-board computation using noisy sensor data.…”