Toward One Million Miles of Pedestrian Detection and Avoidance: Truly Robust Multi-Sensor Safety Systems
Ramanarayan Vasudevan, PhD
Assistant Professor, Mechanical Engineering
May 2016 – April 2018
Matthew Johnson-Roberson, PhD
Assistant Professor, Naval Architecture and Marine Engineering
Construct a multi-sensor fusion based approach for pedestrian and cyclist detection and a real-time optimization based approach for collision avoidance.
We will use multiple sensors to construct an algorithm to detect and predict the motion of pedestrians and cyclists that is robust to occlusions from any single sensing modality. This algorithm will be trained upon a realistic dataset drawn from existing MTC test beds that is orders of magnitude larger than any existing dataset. We will also construct a real-time optimization scheme to generate controls that avoid collisions with the predicted locations generated by the first algorithm to illustrate the utility of the predictions.
- A 10,000-hour annotated dataset of pedestrian and cyclist motion using the Safety Pilot dataset, which includes data from inclement weather.
- A multi-sensor fusion based approach to detect and predict the motion of pedestrians and cyclists.
- A real-time numerical optimization scheme that synthesizes controller interventions, which can safely avoid probabilistic predictions of pedestrian motion.