Multi-sensor Detection and Tracking using Traversability Based Prediction

In order for fully autonomous robots to be integrated effectively into small combat teams, unmanned ground vehicles (UGVs) must maneuver safely and intelligently among military ground forces and non-combatants. To achieve such robust obstacle detection capability in unstructured environments, it is necessary to fuse information from various sensor modalities. Under various programs in the past, our team has developed obstacle detection/tracking capabilities using Ladar, LWIR and a visible light camera. The team created by Robotic Research, LLC (RR) and General Dynamics Robotic Systems (GDRS) provides a unique set of expertise and on the field experience that cannot be matched in the industry. Kalman filters are the most common approach for tracking and prediction of obstacles but are extremely limited in complex environments. The TACTIC prediction system will drastically increase the efficacy of tracking behind occlusions and in highly cluttered terrain. Our team has the exclusive capability to leverage detection algorithms, planning algorithms, simulation environments, and actual robotic vehicles that are required to make this project a success.