Atrisha Sarkar’s research while affiliated with University of Waterloo and other places

What is this page?


This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.

Publications (16)


Revealed Multi-Objective Utility Aggregation in Human Driving
  • Preprint

March 2023

·

6 Reads

Atrisha Sarkar

·

Kate Larson

·

A central design problem in game theoretic analysis is the estimation of the players' utilities. In many real-world interactive situations of human decision making, including human driving, the utilities are multi-objective in nature; therefore, estimating the parameters of aggregation, i.e., mapping of multi-objective utilities to a scalar value, becomes an essential part of game construction. However, estimating this parameter from observational data introduces several challenges due to a host of unobservable factors, including the underlying modality of aggregation and the possibly boundedly rational behaviour model that generated the observation. Based on the concept of rationalisability, we develop algorithms for estimating multi-objective aggregation parameters for two common aggregation methods, weighted and satisficing aggregation, and for both strategic and non-strategic reasoning models. Based on three different datasets, we provide insights into how human drivers aggregate the utilities of safety and progress, as well as the situational dependence of the aggregation process. Additionally, we show that irrespective of the specific solution concept used for solving the games, a data-driven estimation of utility aggregation significantly improves the predictive accuracy of behaviour models with respect to observed human behaviour.


Generalized Dynamic Cognitive Hierarchy Models for Strategic Driving Behavior

June 2022

·

17 Reads

·

3 Citations

Proceedings of the AAAI Conference on Artificial Intelligence

While there has been an increasing focus on the use of game theoretic models for autonomous driving, empirical evidence shows that there are still open questions around dealing with the challenges of common knowledge assumptions as well as modeling bounded rationality. To address some of these practical challenges, we develop a framework of generalized dynamic cognitive hierarchy for both modelling naturalistic human driving behavior as well as behavior planning for autonomous vehicles (AV). This framework is built upon a rich model of level-0 behavior through the use of automata strategies, an interpretable notion of bounded rationality through safety and maneuver satisficing, and a robust response for planning. 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.



Fig. 2: Simple behavior tree for entering crosswalk decision demonstrating the use of condition (diamond), maneuver (ellipse), selector (?), and sequence (→) nodes.
Fig. 3: Multi-layered model diagram with process and information flow
A Hierarchical Pedestrian Behavior Model to Generate Realistic Human Behavior in Traffic Simulation
  • Preprint
  • File available

May 2022

·

45 Reads

Modelling pedestrian behavior is crucial in the development and testing of autonomous vehicles. In this work, we present a hierarchical pedestrian behavior model that generates high-level decisions through the use of behavior trees, in order to produce maneuvers executed by a low-level motion planner using an adapted Social Force model. A full implementation of our work is integrated into GeoScenario Server, a scenario definition and execution engine, extending its vehicle simulation capabilities with pedestrian simulation. The extended environment allows simulating test scenarios involving both vehicles and pedestrians to assist in the scenario-based testing process of autonomous vehicles. The presented hierarchical model is evaluated on two real-world data sets collected at separate locations with different road structures. Our model is shown to replicate the real-world pedestrians' trajectories with a high degree of fidelity and a decision-making accuracy of 98% or better, given only high-level routing information for each pedestrian.

Download


A taxonomy of strategic human interactions in traffic conflicts

September 2021

·

23 Reads

In order to enable autonomous vehicles (AV) to navigate busy traffic situations, in recent years there has been a focus on game-theoretic models for strategic behavior planning in AVs. However, a lack of common taxonomy impedes a broader understanding of the strategies the models generate as well as the development of safety specification to identity what strategies are safe for an AV to execute. Based on common patterns of interaction in traffic conflicts, we develop a taxonomy for strategic interactions along the dimensions of agents' initial response to right-of-way rules and subsequent response to other agents' behavior. Furthermore, we demonstrate a process of automatic mapping of strategies generated by a strategic planner to the categories in the taxonomy, and based on vehicle-vehicle and vehicle-pedestrian interaction simulation, we evaluate two popular solution concepts used in strategic planning in AVs, QLk and Subgame perfect ϵ\epsilon-Nash Equilibrium, with respect to those categories.


I Know You Can't See Me: Dynamic Occlusion-Aware Safety Validation of Strategic Planners for Autonomous Vehicles Using Hypergames

September 2021

·

34 Reads

A particular challenge for both autonomous and human driving is dealing with risk associated with dynamic occlusion, i.e., occlusion caused by other vehicles in traffic. Based on the theory of hypergames, we develop a novel multi-agent dynamic occlusion risk (DOR) measure for assessing situational risk in dynamic occlusion scenarios. Furthermore, we present a white-box, scenario-based, accelerated safety validation framework for assessing safety of strategic planners in AV. Based on evaluation over a large naturalistic database, our proposed validation method achieves a 4000% speedup compared to direct validation on naturalistic data, a more diverse coverage, and ability to generalize beyond the dataset and generate commonly observed dynamic occlusion crashes in traffic in an automated manner.


Generalized dynamic cognitive hierarchy models for strategic driving behavior

September 2021

·

24 Reads

While there has been an increasing focus on the use of game theoretic models for autonomous driving, empirical evidence shows that there are still open questions around dealing with the challenges of common knowledge assumptions as well as modeling bounded rationality. To address some of these practical challenges, we develop a framework of generalized dynamic cognitive hierarchy for both modelling naturalistic human driving behavior as well as behavior planning for autonomous vehicles (AV). This framework is built upon a rich model of level-0 behavior through the use of automata strategies, an interpretable notion of bounded rationality through safety and maneuver satisficing, and a robust response for planning. 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.


Solution Concepts in Hierarchical Games Under Bounded Rationality With Applications to Autonomous Driving

May 2021

·

11 Reads

·

6 Citations

Proceedings of the AAAI Conference on Artificial Intelligence

With autonomous vehicles (AV) set to integrate further into regular human traffic, there is an increasing consensus of treating AV motion planning as a multi-agent problem. However, the traditional game theoretic assumption of complete rationality is too strong for the purpose of human driving, and there is a need for understanding human driving as a bounded rational activity through a behavioral game theoretic lens. To that end, we adapt three metamodels of bounded rational behavior; two based on Quantal level-k and one based on Nash equilibria with quantal errors. We formalize the different solution concepts that can be applied in the context of hierarchical games, a framework used in multi-agent motion planning, for the purpose of creating game theoretic models of driving behavior. Furthermore, based on a contributed dataset of human driving at a busy urban intersection with a total of ~4k agents and ~44k decision points, we evaluate the behavior models on the basis of model fit to naturalistic data, as well as their predictive capacity. Our results suggest that among the behavior models evaluated, modeling driving behavior as pure strategy Nash equilibria with quantal errors at the level of maneuvers with bounds sampling of actions at the level of trajectories provides the best fit to naturalistic driving behavior, and there is a significant impact of situational factors on the performance of behavior models.


Solution Concepts in Hierarchical Games with Applications to Autonomous Driving

September 2020

·

28 Reads

With autonomous vehicles (AV) set to integrate further into regular human traffic, there is an increasing consensus of treating AV motion planning as a multi-agent problem. However, the traditional game theoretic assumption of complete rationality is too strong for the purpose of human driving, and there is a need for understanding human driving as a bounded rational activity through a behavioral game theoretic lens. To that end, we adapt three metamodels of bounded rational behavior; two based on Quantal level-k and one based on Nash equilibrium with quantal errors. We formalize the different solution concepts that can be applied in the context of hierarchical games, a framework used in multi-agent motion planning, for the purpose of creating game theoretic models of driving behavior. Furthermore, based on a contributed dataset of human driving at a busy urban intersection with a total of ~4k agents and ~44k decision points, we evaluate the behavior models on the basis of model fit to naturalistic data, as well as their predictive capacity. Our results suggest that among the behavior models evaluated, modeling driving behavior as pure strategy NE with quantal errors at the level of maneuvers with bounds sampling of actions at the level of trajectories provides the best fit to naturalistic driving behavior.


Citations (8)


... Data-driven deep learning methods must rely on massive big data sample training (accumulated actual driving miles need to reach tens of billions of miles) to deal with the "long tail" problem that may cause driving accidents 11,12 and the explosive progress of autonomous driving technology being hindered. Daily, low probability of occurrence or emergent, dangerous, and the edge (corner) scene of scarce samples are often related to irrational behavior and interaction 13,14 . And pure deep learning method based on data driven is difficult to effectively cope with the "long tail" problem 15 . ...

Reference:

Quantum decision making in automatic driving
Solution Concepts in Hierarchical Games Under Bounded Rationality With Applications to Autonomous Driving
  • Citing Article
  • May 2021

Proceedings of the AAAI Conference on Artificial Intelligence

... However, it lacks a trajectory planner, making it impossible to plan flexible and realistic trajectories. Larter et al. [32] utilize behavior trees to control pedestrians (i.e., setting motion objectives) in simulation. As far as we know, no work combines behavior trees with the maneuver decisions of NPC vehicles to create an adversarial scenario in ADS simulation testing. ...

A Hierarchical Pedestrian Behavior Model to Generate Realistic Human Behavior in Traffic Simulation
  • Citing Conference Paper
  • June 2022

... We will also expand the behavior trees and maneuvers for interaction with pedestrians [44]. Finally, we plan to exploit the model in generating new scenarios by injecting road-user misbehaviors into behavior trees, such as simulating distraction [59] and ignoring occlusions [60]. The SDV model implementation and toolset to design and run scenarios is publicly available and can be integrated with any simulation environment via co-simulation. ...

I Know You Can't See Me: Dynamic Occlusion-Aware Safety Validation of Strategic Planners for Autonomous Vehicles Using Hypergames
  • Citing Conference Paper
  • May 2022

... Accurate perception of surrounding vehicle driving behavior and anticipation of possible dangers can significantly reduce traffic accidents caused by lane changes during high-speed driving, improve the decision-making capabilities of intelligent vehicles, and enhance driving comfort and safety [5,6]. Studies on vehicle lane change intention can be broadly categorized into two types: rulebased approaches [7,8] and model-based approaches [9][10][11][12]. ...

Generalized Dynamic Cognitive Hierarchy Models for Strategic Driving Behavior
  • Citing Article
  • June 2022

Proceedings of the AAAI Conference on Artificial Intelligence

... Chen et al. proposed a method that integrates frame intervals and lateral displacement errors in trajectory data to identify the starting and ending points of lane-changing maneuvers [45]. In addition, there are currently some limitations to the model [46,47]. ...

A behavior driven approach for sampling rare event situations for autonomous vehicles
  • Citing Conference Paper
  • November 2019

... To mention a few, Feature Models in the Wild (LVAT) (Berger et al., 2013) accurate world feature models in the operating systems domain, including large-scale feature models such as the Linux kernel, with more than 6000 features, and Ecos, with more than 1000 features. These feature models were initially defined in the Linux KConfig language and translated into the Clafer language (Juodisius et al., 2019). More recently, Knüppel et al. (2017b) provide newer versions of these feature models used for evaluation, 21 and they are available in UVL due to the integration of UVL into the FeatureIDE tool (Sundermann et al., 2021b). ...

Clafer: Lightweight Modeling of Structure, Behaviour, and Variability

The Art Science and Engineering of Programming

... Furthermore, results show [30] the screen-based exposure to models may improve comprehension time and induce fewer comprehension errors than reading models on paper. Similarly, Zayan et al. [31] showed in an experiment involving 26 graduate students that participants were up to 80% less likely to make mistakes during model comprehension when provided with illustrative examples and asked up to 90% fewer clarification questions to domain experts. ...

Example-driven modeling: on effects of using examples on structural model comprehension, what makes them useful, and how to create them

Software and Systems Modeling

... The above, together with the other vast majority of work on configuration performance modeling that solely seeks to improve model accuracy [24], [34], [37], [39], [40], [80], implies that the community tends to believe the accuracy is the key that strongly impacts whether mode-based tuners should be favored over their model-free counterparts (i.e., 20 Nevertheless, despite the widely followed general belief, we have never had a thorough understanding of how the models impact the tuning process. In fact, a few preliminary works [69], [70] and our years of experience have hinted that the above belief can be inaccurate or even misleading. ...

Data-efficient performance learning for configurable systems

Empirical Software Engineering