Comparing merging behaviors observed in naturalistic data with behaviors generated by a machine learned model

There is quickly growing literature on machine-learned models that predict human driving trajectories in road traffic. These models focus their learning on low-dimensional error metrics, for example average distance between model-generated and observed trajectories. Such metrics permit relative comparison of models, but do not provide clearly interpretable information on how close to human behavior the models actually come, for example in terms of higher-level behavior phenomena that are known to be present in human driving. We study highway driving as an example scenario, and introduce metrics to quantitatively demonstrate the presence, in a naturalistic dataset, of two familiar behavioral phenomena: (1) The kinematics-dependent contest, between on-highway and on-ramp vehicles, of who passes the merging point first. (2) Courtesy lane changes away from the outermost lane, to leave space for a merging vehicle. Applying the exact same metrics to the output of a state-of-the-art machine-learned model, we show that the model is capable of reproducing the former phenomenon, but not the latter. We argue that this type of behavioral analysis provides information that is not available from conventional model-fitting metrics, and that it may be useful to analyze (and possibly fit) models also based on these types of behavioral criteria.

[1]  Matthias Althoff,et al.  Pedestrian Models for Autonomous Driving Part I: Low-Level Models, From Sensing to Tracking , 2020, IEEE Transactions on Intelligent Transportation Systems.

[2]  Alexandra Kondyli,et al.  Driver Behavior at Freeway-Ramp Merging Areas , 2009 .

[3]  Natasha Merat,et al.  Defining interactions: a conceptual framework for understanding interactive behaviour in human and automated road traffic , 2020 .

[4]  Mooi Choo Chuah,et al.  GRIP: Graph-based Interaction-aware Trajectory Prediction , 2019, 2019 IEEE Intelligent Transportation Systems Conference (ITSC).

[5]  Zuduo Zheng,et al.  Recent developments and research needs in modeling lane changing , 2014 .

[6]  Gustav Markkula,et al.  A farewell to brake reaction times? Kinematics-dependent brake response in naturalistic rear-end emergencies. , 2016, Accident; analysis and prevention.

[7]  Mohan M. Trivedi,et al.  Convolutional Social Pooling for Vehicle Trajectory Prediction , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[8]  David N. Lee,et al.  A Theory of Visual Control of Braking Based on Information about Time-to-Collision , 1976, Perception.

[9]  Catherine M. Burns,et al.  Autonomous Driving in the Real World: Experiences with Tesla Autopilot and Summon , 2016, AutomotiveUI.

[10]  Wataru Ike,et al.  Pre-Collision System for Toyota Safety Sense , 2016 .

[11]  Dinesh Manocha,et al.  TrafficPredict: Trajectory Prediction for Heterogeneous Traffic-Agents , 2018, AAAI.

[12]  Matthias Althoff,et al.  Pedestrian Models for Autonomous Driving Part II: High-Level Models of Human Behavior , 2020, IEEE Transactions on Intelligent Transportation Systems.

[13]  Moshe Ben-Akiva,et al.  Modeling Acceleration Decisions for Freeway Merges , 2009 .

[14]  Hesham Rakha,et al.  Game Theoretical Approach to Model Decision Making for Merging Maneuvers at Freeway On-Ramps , 2017 .

[15]  Omar Y. Al-Jarrah,et al.  Deep Learning-based Vehicle Behaviour Prediction For Autonomous Driving Applications: A Review , 2019, ArXiv.

[16]  Silvio Savarese,et al.  Social LSTM: Human Trajectory Prediction in Crowded Spaces , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[17]  Reza Langari,et al.  A human-like game theory-based controller for automatic lane changing , 2018 .

[18]  Dominique Gruyer,et al.  A Review of Motion Planning for Highway Autonomous Driving , 2020, IEEE Transactions on Intelligent Transportation Systems.