Predicting Pedestrian Trajectories Using Velocity-Space Reasoning

We introduce a novel method to predict pedestrian trajectories using agent-based velocity-space reasoning for improved human-robot interaction. This formulation models the trajectory of each moving pedestrian in a robot’s environment using velocity obstacles and learns the simulation parameters based on tracked data. The resulting motion model for each agent is computed using statistical inferencing techniques from noisy data. This includes the combination of Ensemble Kalman filters and maximum likelihood estimation algorithm to learn individual motion parameters at interactive rates. We highlight the performance of our motion model in real-world crowded scenarios. We also compare its performance with prior techniques and demonstrate improved accuracy in the predicted trajectories.

[1]  Craig W. Reynolds.  Steering Behaviors For Autonomous Characters , 1999 .

[2]  Dinesh Manocha,et al.  PLEdestrians: a least-effort approach to crowd simulation , 2010, SCA '10.

[3]  Helbing,et al.  Social force model for pedestrian dynamics. , 1995, Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics.

[4]  G. McLachlan,et al.  The EM algorithm and extensions , 1996 .

[5]  Kai Oliver Arras,et al.  People tracking with human motion predictions from social forces , 2010, 2010 IEEE International Conference on Robotics and Automation.

[6]  Hubert Klüpfel,et al.  Evacuation Dynamics: Empirical Results, Modeling and Applications , 2009, Encyclopedia of Complexity and Systems Science.

[7]  Wolfram Burgard,et al.  Learning Motion Patterns of People for Compliant Robot Motion , 2005, Int. J. Robotics Res..

[8]  Andreas Schadschneider,et al.  Automatic Extraction of Pedestrian Trajectories from Video Recordings , 2010 .

[9]  Norman I. Badler,et al.  Virtual Crowds: Methods, Simulation, and Control , 2008, Virtual Crowds: Methods, Simulation, and Control.

[10]  Mark H. Overmars,et al.  A Predictive Collision Avoidance Model for Pedestrian Simulation , 2009, MIG.

[11]  Siddhartha S. Srinivasa,et al.  Planning-based prediction for pedestrians , 2009, 2009 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[12]  Stéphane Donikian,et al.  Experiment-based modeling, simulation and validation of interactions between virtual walkers , 2009, SCA '09.

[13]  Henry A. Kautz,et al.  Voronoi tracking: location estimation using sparse and noisy sensor data , 2003, Proceedings 2003 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2003) (Cat. No.03CH37453).

[14]  Luc Van Gool,et al.  You'll never walk alone: Modeling social behavior for multi-target tracking , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[15]  Jianbo Shi,et al.  Multi-hypothesis motion planning for visual object tracking , 2011, 2011 International Conference on Computer Vision.

[16]  Maja J. Mataric,et al.  A laser-based people tracker , 2002, Proceedings 2002 IEEE International Conference on Robotics and Automation (Cat. No.02CH37292).

[17]  Dinesh Manocha,et al.  Reciprocal Velocity Obstacles for real-time multi-agent navigation , 2008, 2008 IEEE International Conference on Robotics and Automation.

[18]  Ramin Mehran,et al.  Abnormal crowd behavior detection using social force model , 2009, CVPR.

[19]  Dinesh Manocha,et al.  Reciprocal n-Body Collision Avoidance , 2011, ISRR.

[20]  Daniel Cohen-Or,et al.  Fitting behaviors to pedestrian simulations , 2009, SCA '09.

[21]  Mark H. Overmars,et al.  A Velocity-Based Approach for Simulating Human Collision Avoidance , 2010, IVA.

[22]  Takeo Kanade,et al.  Tracking in unstructured crowded scenes , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[23]  Wolfram Burgard,et al.  People Tracking with Mobile Robots Using Sample-Based Joint Probabilistic Data Association Filters , 2003, Int. J. Robotics Res..

[24]  Adrien Treuille,et al.  Continuum crowds , 2006, SIGGRAPH 2006.

[25]  Dinesh Manocha,et al.  Simulating heterogeneous crowd behaviors using personality trait theory , 2011, SCA '11.

[26]  Ryosuke Shibasaki,et al.  Tracking multiple people using laser and vision , 2005, 2005 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[27]  Luis E. Ortiz,et al.  Who are you with and where are you going? , 2011, CVPR 2011.

[28]  Geoffrey J. Gordon,et al.  Better Motion Prediction for People-tracking , 2004 .

[29]  Ko Nishino,et al.  Tracking Pedestrians Using Local Spatio-Temporal Motion Patterns in Extremely Crowded Scenes , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.