Symphony: Learning Realistic and Diverse Agents for Autonomous Driving Simulation

Igl, Maximilian; Kim, Dae Woo; Kuefler, Alex; Mougin, Paul; Shah, Punit; Shiarlis, Kyriacos; Anguelov, Dragomir; Palatucci, Mark; White, Brandyn; Whiteson, Shimon

doi:10.1109/icra46639.2022.9811990

Cited by 21 publications

(5 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To this end, methods that use constraints online in the form of Control Barrier Functions [17] and one-step state feedback [9] have been proposed, but neither uses expert demonstrations. [18] takes an alternate approach to improving efficiency and applicability of LfD-generated policies using beam search with goal generation as a hierarchical approach. The closest to our work is the recently proposed method [16] that uses STL and expert demonstrations to synthesize a trajectoryfeedback controller.…”

Section: Related Workmentioning

confidence: 99%

Follow The Rules: Online Signal Temporal Logic Tree Search for Guided Imitation Learning in Stochastic Domains

Jerry¹,

Scherer²,

Parv³

et al. 2022

Preprint

View full text Add to dashboard Cite

Seamlessly integrating rules in Learning-from-Demonstrations (LfD) policies is a critical requirement to enable the real-world deployment of AI agents. Recently Signal Temporal Logic (STL) has been shown to be an effective language for encoding rules as spatio-temporal constraints. This work uses Monte Carlo Tree Search (MCTS) as a means of integrating STL specification into a vanilla LfD policy to improve constraint satisfaction. We propose augmenting the MCTS heuristic with STL robustness values to bias the tree search towards branches with higher constraint satisfaction. While the domain-independent method can be applied to integrate STL rules online into any pre-trained LfD algorithm, we choose goalconditioned Generative Adversarial Imitation Learning as the offline LfD policy. We apply the proposed method to the domain of planning trajectories for General Aviation aircraft around a non-towered airfield. Results using the simulator trained on real-world data showcase 60% improved performance over baseline LfD methods that do not use STL heuristics. [Code] 1 [Video] 2 *equal contribution This work was supported by Army Futures Command.

show abstract

Section: Related Workmentioning

confidence: 99%

Follow The Rules: Online Signal Temporal Logic Tree Search for Guided Imitation Learning in Stochastic Domains

Jerry¹,

Scherer²,

Parv³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Especially, we aim to produce safety-critical events with distribution-level accuracy, including both crashes and near-misses, which are critical for training and testing AVs. This differentiates our proposed NeuralNDE model from most existing simulators based on imitation learning (including generative adversarial imitation learning) [30][31][32][33][34][35][36] , where statistical realism is hardly considered and cannot be achieved. For example, the crash rates of these simulation environments are significantly higher (e.g., SimNet 33 ) than that of real-world traffic.…”

mentioning

confidence: 94%

Learning naturalistic driving environment with statistical realism

et al. 2023

View full text Add to dashboard Cite

For simulation to be an effective tool for the development and testing of autonomous vehicles, the simulator must be able to produce realistic safety-critical scenarios with distribution-level accuracy. However, due to the high dimensionality of real-world driving environments and the rarity of long-tail safety-critical events, how to achieve statistical realism in simulation is a long-standing problem. In this paper, we develop NeuralNDE, a deep learning-based framework to learn multi-agent interaction behavior from vehicle trajectory data, and propose a conflict critic model and a safety mapping network to refine the generation process of safety-critical events, following real-world occurring frequencies and patterns. The results show that NeuralNDE can achieve both accurate safety-critical driving statistics (e.g., crash rate/type/severity and near-miss statistics, etc.) and normal driving statistics (e.g., vehicle speed/distance/yielding behavior distributions, etc.), as demonstrated in the simulation of urban driving environments. To the best of our knowledge, this is the first time that a simulation model can reproduce the real-world driving environment with statistical realism, particularly for safety-critical situations.

show abstract

“…Contemporary generative AI models can learn complex driving behaviors from large quantities of data (e.g., Suo et al, 2021 ; Igl et al, 2022 ). There has also been extensive work in machine learning and robotics on developing models able to manage uncertainty through concepts like artificial curiosity and intrinsic motivation (e.g., Schmidhuber, 1991 ; Sun et al, 2011 ; Hester and Stone, 2012 ).…”

Section: Introductionmentioning

confidence: 99%

Resolving uncertainty on the fly: modeling adaptive driving behavior as active inference

Engström,

Wei,

McDonald

et al. 2024

Front. Neurorobot.

View full text Add to dashboard Cite

Understanding adaptive human driving behavior, in particular how drivers manage uncertainty, is of key importance for developing simulated human driver models that can be used in the evaluation and development of autonomous vehicles. However, existing traffic psychology models of adaptive driving behavior either lack computational rigor or only address specific scenarios and/or behavioral phenomena. While models developed in the fields of machine learning and robotics can effectively learn adaptive driving behavior from data, due to their black box nature, they offer little or no explanation of the mechanisms underlying the adaptive behavior. Thus, generalizable, interpretable, computational models of adaptive human driving behavior are still rare. This paper proposes such a model based on active inference, a behavioral modeling framework originating in computational neuroscience. The model offers a principled solution to how humans trade progress against caution through policy selection based on the single mandate to minimize expected free energy. This casts goal-seeking and information-seeking (uncertainty-resolving) behavior under a single objective function, allowing the model to seamlessly resolve uncertainty as a means to obtain its goals. We apply the model in two apparently disparate driving scenarios that require managing uncertainty, (1) driving past an occluding object and (2) visual time-sharing between driving and a secondary task, and show how human-like adaptive driving behavior emerges from the single principle of expected free energy minimization.

show abstract

Symphony: Learning Realistic and Diverse Agents for Autonomous Driving Simulation

Cited by 21 publications

References 15 publications

Follow The Rules: Online Signal Temporal Logic Tree Search for Guided Imitation Learning in Stochastic Domains

Follow The Rules: Online Signal Temporal Logic Tree Search for Guided Imitation Learning in Stochastic Domains

Learning naturalistic driving environment with statistical realism

Resolving uncertainty on the fly: modeling adaptive driving behavior as active inference

Contact Info

Product

Resources

About