High-Speed Robot Navigation using Predicted Occupancy Maps
This work provides an incremental improvement for high-speed robot navigation, specifically for systems limited by sensor field of view and computational mapping bottlenecks.
The paper addresses the challenge of high-speed robot navigation by predicting occupancy maps beyond the sensor's field of view using a generative neural network. This approach enabled a physical robot to achieve improved navigation performance at 4 m/s compared to a baseline without predicted regions.
Safe and high-speed navigation is a key enabling capability for real world deployment of robotic systems. A significant limitation of existing approaches is the computational bottleneck associated with explicit mapping and the limited field of view (FOV) of existing sensor technologies. In this paper, we study algorithmic approaches that allow the robot to predict spaces extending beyond the sensor horizon for robust planning at high speeds. We accomplish this using a generative neural network trained from real-world data without requiring human annotated labels. Further, we extend our existing control algorithms to support leveraging the predicted spaces to improve collision-free planning and navigation at high speeds. Our experiments are conducted on a physical robot based on the MIT race car using an RGBD sensor where were able to demonstrate improved performance at 4 m/s compared to a controller not operating on predicted regions of the map.