Generalizable intention prediction of human drivers at intersections

Effective navigation of urban environments is a primary challenge remaining in the development of autonomous vehicles. Intersections come in many shapes and forms, making it difficult to find features and models that generalize across intersection types. New and traditional features are used to train several intersection intention models on real-world intersection data, and a new class of recurrent neural networks, Long Short Term Memory networks (LSTMs), are shown to outperform the state of the art. The models predict whether a driver will turn left, turn right, or continue straight up to 150 m with consistent accuracy before reaching the intersection. The results show promise for further use of LSTMs, with the mean cross validated prediction accuracy averaging over 85% for both three and four-way intersections, obtaining 83% for the highest throughput intersection.

[1]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[2]  Richard Hans Robert Hahnloser,et al.  Digital selection and analogue amplification coexist in a cortex-inspired silicon circuit , 2000, Nature.

[3]  Nir Friedman,et al.  Probabilistic Graphical Models - Principles and Techniques , 2009 .

[4]  Yoram Singer,et al.  Adaptive Subgradient Methods for Online Learning and Stochastic Optimization , 2011, J. Mach. Learn. Res..

[5]  Panos E. Trahanias,et al.  Probabilistic Autonomous Robot Navigation in Dynamic Environments with Human Motion Prediction , 2010, Int. J. Soc. Robotics.

[6]  Jonathan P. How,et al.  Threat assessment design for driver assistance system at intersections , 2010, 13th International IEEE Conference on Intelligent Transportation Systems.

[7]  Amaury Nègre,et al.  Probabilistic Analysis of Dynamic Scenes and Collision Risks Assessment to Improve Driving Safety , 2011, IEEE Intelligent Transportation Systems Magazine.

[8]  Christian Laugier,et al.  Evaluating risk at road intersections by detecting conflicting intentions , 2012, 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[9]  Rüdiger Dillmann,et al.  Learning context sensitive behavior models from observations for predicting traffic situations , 2013, 16th International IEEE Conference on Intelligent Transportation Systems (ITSC 2013).

[10]  Kian Hsiang Low,et al.  Interactive POMDP Lite: Towards Practical Planning to Predict and Exploit Intentions for Interacting with Self-Interested Agents , 2013, IJCAI.

[11]  Jonas Firl,et al.  Online maneuver recognition and multimodal trajectory prediction for intersection assistance using non-parametric regression , 2014, 2014 IEEE Intelligent Vehicles Symposium Proceedings.

[12]  Karl-Heinz Hoffmann,et al.  Prediction of driver intended path at intersections , 2014, 2014 IEEE Intelligent Vehicles Symposium Proceedings.

[13]  Volkan Sezer,et al.  Towards autonomous navigation of unsignalized intersections under uncertainty of human driver intent , 2015, 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[14]  Ralph Bruder,et al.  Algorithm for Driver Intention Detection with Fuzzy Logic and Edit Distance , 2015, 2015 IEEE 18th International Conference on Intelligent Transportation Systems.

[15]  Rakesh Gupta,et al.  Turn prediction at generalized intersections , 2015, 2015 IEEE Intelligent Vehicles Symposium (IV).

[16]  Wolfram Burgard,et al.  Socially compliant mobile robot navigation via inverse reinforcement learning , 2016, Int. J. Robotics Res..

[17]  Martín Abadi,et al.  TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems , 2016, ArXiv.

[18]  Mykel J. Kochenderfer,et al.  Analysis of Recurrent Neural Networks for Probabilistic Modeling of Driver Behavior , 2017, IEEE Transactions on Intelligent Transportation Systems.