Abstract
Accurate motion models are key to many tasks in the intelligent vehicle domain, but simple Linear Dynamics (e.g. Kalman filtering) do not exploit the spatio-temporal context of motion. We present a method to learn Switching Linear Dynamics of object tracks observed from within a driving vehicle. Each switching state captures object dynamics as a mean motion with variance, but also has an additional spatial distribution on where the dynamic is seen relative to the vehicle. Thus, both an object's previous movements and current location will make certain dynamics more probable for subsequent time steps. To train the model, we use Bayesian inference to sample parameters from the posterior, and jointly learn the required number of dynamics. Unlike Maximum Likelihood learning, inference is robust against overfitting and poor initialization. We demonstrate our approach on an ego-motion compensated track dataset of pedestrians, and illustrate how the switching dynamics can make more accurate path predictions than a mixture of linear dynamics for crossing pedestrians.
Original language | English |
---|---|
Title of host publication | 2014 IEEE Intelligent Vehicles Symposium, IV 2004 - Proceedings |
Publisher | IEEE |
Pages | 1445-1450 |
Number of pages | 6 |
ISBN (Print) | 9781479936380 |
DOIs | |
Publication status | Published - 2014 |
Externally published | Yes |
Event | 25th IEEE Intelligent Vehicles Symposium, IV 2014 - Dearborn, MI, United States Duration: 8 Jun 2014 → 11 Jun 2014 |
Conference
Conference | 25th IEEE Intelligent Vehicles Symposium, IV 2014 |
---|---|
Country/Territory | United States |
City | Dearborn, MI |
Period | 8/06/14 → 11/06/14 |