High-Speed Robot Navigation using Predicted Occupancy Maps

Abstract

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.

Post-Stall Navigation with Fixed-Wing UAVs using Onboard Vision

“…For instance, by augmenting the optimizer's cost function, we may be able to reduce failure cases by rewarding information gathering during aggressive maneuvers. We may also be able to improve performance by using generative adversarial networks, as was done in [30], to predict and reason about the environment beyond the field of view.…”