Generalized Dynamic Cognitive Hierarchy Models for Strategic Driving Behavior

Authors: Atrisha Sarkar, Kate Larson, Krzysztof Czarnecki5173-5182

AAAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Based on evaluation on two large naturalistic datasets as well as simulation of critical traffic scenarios, we show that i) automata strategies are well suited for level-0 behavior in a dynamic level-k framework, and ii) the proposed robust response to a heterogeneous population of strategic and non-strategic reasoners can be an effective approach for game theoretic planning in AV.
Researcher Affiliation Academia Atrisha Sarkar 1, Kate Larson 1, Krzysztof Czarnecki 2 1David R Cheriton School of Computer Science 2Department of Electrical and Computer Engineering University of Waterloo, Ontario, Canada
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks.
Open Source Code Yes Code and supplementary videos are available at https://git.uwaterloo.ca/a9sarkar/repeated-driving-games.
Open Datasets Yes First, we compare the models with respect to large naturalistic observational driving data using a) the Intersection dataset from the Waterloo multi-agent traffic dataset (WMA) recorded at a busy Canadian intersection (Sarkar and Czarnecki 2021) available at http://wiselab.uwaterloo.ca/waterloo-multi-agent-trafficdataset/, and b) the in D dataset recorded at intersections in Germany (Bock et al. 2020) (Fig. 3a).
Dataset Splits No The paper evaluates model performance based on 'match rate' over the entire dataset or simulations, but does not provide specific training/validation/test dataset splits needed for reproducibility of a machine learning model training process.
Hardware Specification No The paper does not provide specific hardware details (e.g., exact GPU/CPU models, memory amounts) used for running its experiments.
Software Dependencies No The paper does not provide specific ancillary software details, such as library names with version numbers.
Experiment Setup Yes From both datasets, which include around 10k vehicles in total, we extract the long duration unprotected left turn (LT) and right turn (RT) scenarios, and instantiate games between left (and right) turning vehicles and oncoming vehicles with th = 6s and tp = 2s, resulting in a total of 1678 games. The second part of the evaluation is based on simulation of three critical traffic scenarios derived from the NHTSA precrash database (Najm et al. 2007), where we instantiate agents with a range of safety aspirations as well as initial game states, and evaluate the outcome of the game based on each model. All the games in the experiments are 2 agent games with the exception of one of the simulation of critical scenario (intersection clearance), which is a 3 agent game. We use the same parameters used in (Tian et al. 2021) for the precision parameters in the QLk model.