FURI | Spring 2023
Social Navigation for Autonomous Vehicles
The main objective of this research project was to develop a computationally efficient motion planner that optimizes task efficiency and is collision averse. The motion planner incorporates 2D vehicle dynamics and produces trajectories that mirror human driving patterns. The accuracy and efficiency of the developed motion planner were evaluated through an interactive human driver dataset (the INTERACTION Dataset). The developed motion planner optimizes safety by reducing the risk associated with operating a vehicle, and by lessening the likelihood of a vehicle collision due to human error. Future work includes extending the motion planner to utilize empathetic intent inference to more accurately gauge the intended trajectories of surrounding vehicles.
Hometown: San Diego, California, United States
Graduation date: Spring 2023