All Weather Perception: Joint Data Association, Tracking, and Classification for Autonomous Ground Vehicles
This work addresses perception reliability for autonomous vehicles in adverse weather, though it appears incremental as it extends existing methods with classification and sensor integration.
The authors tackled the problem of robust perception for autonomous ground vehicles in all-weather conditions by developing a joint probabilistic algorithm for data association, tracking, and classification, resulting in experimental demonstration of complementary sensor fusion using camera, lidar, and radar.
A novel probabilistic perception algorithm is presented as a real-time joint solution to data association, object tracking, and object classification for an autonomous ground vehicle in all-weather conditions. The presented algorithm extends a Rao-Blackwellized Particle Filter originally built with a particle filter for data association and a Kalman filter for multi-object tracking (Miller et al. 2011a) to now also include multiple model tracking for classification. Additionally a state-of-the-art vision detection algorithm that includes heading information for autonomous ground vehicle (AGV) applications was implemented. Cornell's AGV from the DARPA Urban Challenge was upgraded and used to experimentally examine if and how state-of-the-art vision algorithms can complement or replace lidar and radar sensors. Sensor and algorithm performance in adverse weather and lighting conditions is tested. Experimental evaluation demonstrates robust all-weather data association, tracking, and classification where camera, lidar, and radar sensors complement each other inside the joint probabilistic perception algorithm.