Core Concepts
AMCO integrates vision and proprioception for stable robot navigation in challenging terrains.
Abstract
AMCO combines vision-based and proprioception-based perception capabilities for quadruped robot navigation.
Three cost maps are used: general knowledge map, traversability history map, and current proprioception map.
The adaptive coupling of these maps ensures reliable navigation decisions based on the most accurate input modality.
A novel planner selects appropriate gaits and velocities, resulting in improved stability and success rates.
Performance evaluation shows significant reductions in IMU energy density and improvements in navigation success rate compared to other methods.
Stats
"We demonstrate AMCO’s navigation performance in different real-world outdoor environments and observe 10.8%-34.9% reduction w.r.t. two stability metrics."
"Up to 50% improvement in terms of success rate compared to current navigation methods."