Autonomous Navigation in Dynamic Environments\ using Sensor-Fusion based Multiple Object Tracking

J. Patel

To navigate safely through crowded dynamic environments, an autonomous system must be capable of planning a path while incorporating the behavior of surrounding objects. GET Lab currently employs a trajectory planner in conjunction with a reactive obstacle avoidance method. However, the behavior of dynamic objects is not taken into consideration. To address this, Multiple Object Tracking (MOT) will be used in this thesis to create a risk map that builds a model of potentially safe or unsafe areas in the robot's environment. To plan a safe path to the goal, this risk map has to be integrated into the navigation system, while taking real-time conditions into account. For evaluation, the system will be tested on simulated and real world environments.