David Isele

Mohamed Elnoor, Francesca Baldini, Ananya Trivedi et al.

Motorcycles face disproportionately high crash risks compared to cars due to limited protection and heightened sensitivity to surface hazards, yet Advanced Rider Assistance Systems (ARAS) remain underdeveloped relative to Advanced Driver Assistance Systems (ADAS). We propose a novel ARAS that enhances motorcycle safety through semantic perception and risk-aware planning. Our approach leverages Vision-Language Models (VLMs) for contextual hazard reasoning and integrates them with segmentation-based detection to construct dense risk maps. These maps encode both semantic characteristics (e.g., pothole severity, puddle slipperiness) and physical attributes (e.g., size, depth), which produce per-pixel hazard costs that capture motorcycle-specific risks. These maps are used by a sampling-based planner tailored to motorcycle dynamics to recommend throttle and steering actions that minimize hazard exposure while advancing toward the destination. We evaluate our system in different scenarios in the CARLA simulator. Compared to the baseline method, our method achieves higher success rates and lower hazard exposure, while qualitative results demonstrate interpretable risk maps and safe trajectory recommendations.

Jiachen Li, David Isele, Kanghoon Lee et al.

Deep reinforcement learning (DRL) provides a promising way for intelligent agents (e.g., autonomous vehicles) to learn to navigate complex scenarios. However, DRL with neural networks as function approximators is typically considered a black box with little explainability and often suffers from suboptimal performance, especially for autonomous navigation in highly interactive multi-agent environments. To address these issues, we propose three auxiliary tasks with spatio-temporal relational reasoning and integrate them into the standard DRL framework, which improves the decision making performance and provides explainable intermediate indicators. We propose to explicitly infer the internal states (i.e., traits and intentions) of surrounding agents (e.g., human drivers) as well as to predict their future trajectories in the situations with and without the ego agent through counterfactual reasoning. These auxiliary tasks provide additional supervision signals to infer the behavior patterns of other interactive agents. Multiple variants of framework integration strategies are compared. We also employ a spatio-temporal graph neural network to encode relations between dynamic entities, which enhances both internal state inference and decision making of the ego agent. Moreover, we propose an interactivity estimation mechanism based on the difference between predicted trajectories in these two situations, which indicates the degree of influence of the ego agent on other agents. To validate the proposed method, we design an intersection driving simulator based on the Intelligent Intersection Driver Model (IIDM) that simulates vehicles and pedestrians. Our approach achieves robust and state-of-the-art performance in terms of standard evaluation metrics and provides explainable intermediate indicators (i.e., internal states, and interactivity scores) for decision making.

Haimin Hu, David Isele, Sangjae Bae et al.

The ability to accurately predict others' behavior is central to the safety and efficiency of interactive robotics. Unfortunately, robots often lack access to key information on which these predictions may hinge, such as other agents' goals, attention, and willingness to cooperate. Dual control theory addresses this challenge by treating unknown parameters of a predictive model as stochastic hidden states and inferring their values at runtime using information gathered during system operation. While able to optimally and automatically trade off exploration and exploitation, dual control is computationally intractable for general interactive motion planning. In this paper, we present a novel algorithmic approach to enable active uncertainty reduction for interactive motion planning based on the implicit dual control paradigm. Our approach relies on sampling-based approximation of stochastic dynamic programming, leading to a model predictive control problem that can be readily solved by real-time gradient-based optimization methods. The resulting policy is shown to preserve the dual control effect for a broad class of predictive models with both continuous and categorical uncertainty. To ensure the safe operation of the interacting agents, we use a runtime safety filter (also referred to as a "shielding" scheme), which overrides the robot's dual control policy with a safety fallback strategy when a safety-critical event is imminent. We then augment the dual control framework with an improved variant of the recently proposed shielding-aware robust planning scheme, which proactively balances the nominal planning performance with the risk of high-cost emergency maneuvers triggered by low-probability agent behaviors. We demonstrate the efficacy of our approach with both simulated driving studies and hardware experiments using 1/10 scale autonomous vehicles.

Kanghoon Lee, Jiachen Li, David Isele et al.

Although deep reinforcement learning (DRL) has shown promising results for autonomous navigation in interactive traffic scenarios, existing work typically adopts a fixed behavior policy to control social vehicles in the training environment. This may cause the learned driving policy to overfit the environment, making it difficult to interact well with vehicles with different, unseen behaviors. In this work, we introduce an efficient method to train diverse driving policies for social vehicles as a single meta-policy. By randomizing the interaction-based reward functions of social vehicles, we can generate diverse objectives and efficiently train the meta-policy through guiding policies that achieve specific objectives. We further propose a training strategy to enhance the robustness of the ego vehicle's driving policy using the environment where social vehicles are controlled by the learned meta-policy. Our method successfully learns an ego driving policy that generalizes well to unseen situations with out-of-distribution (OOD) social agents' behaviors in a challenging uncontrolled T-intersection scenario.

Xiaobai Ma, David Isele, Jayesh K. Gupta et al.

Multi-agent reinforcement learning (MARL) provides an efficient way for simultaneously learning policies for multiple agents interacting with each other. However, in scenarios requiring complex interactions, existing algorithms can suffer from an inability to accurately anticipate the influence of self-actions on other agents. Incorporating an ability to reason about other agents' potential responses can allow an agent to formulate more effective strategies. This paper adopts a recursive reasoning model in a centralized-training-decentralized-execution framework to help learning agents better cooperate with or compete against others. The proposed algorithm, referred to as the Recursive Reasoning Graph (R2G), shows state-of-the-art performance on multiple multi-agent particle and robotics games.

Mansur Arief, Mike Timmerman, Jiachen Li et al.

Training intelligent agents to navigate highly interactive environments presents significant challenges. While guided meta reinforcement learning (RL) approach that first trains a guiding policy to train the ego agent has proven effective in improving generalizability across scenarios with various levels of interaction, the state-of-the-art method tends to be overly sensitive to extreme cases, impairing the agents' performance in the more common scenarios. This study introduces a novel training framework that integrates guided meta RL with importance sampling (IS) to optimize training distributions iteratively for navigating highly interactive driving scenarios, such as T-intersections or roundabouts. Unlike traditional methods that may underrepresent critical interactions or overemphasize extreme cases during training, our approach strategically adjusts the training distribution towards more challenging driving behaviors using IS proposal distributions and applies the importance ratio to de-bias the result. By estimating a naturalistic distribution from real-world datasets and employing a mixture model for iterative training refinements, the framework ensures a balanced focus across common and extreme driving scenarios. Experiments conducted with both synthetic and naturalistic datasets demonstrate both accelerated training and performance improvements under highly interactive driving tasks.

Jinning Li, Jiachen Li, Sangjae Bae et al.

Deep learning-based trajectory prediction models for autonomous driving often struggle with generalization to out-of-distribution (OOD) scenarios, sometimes performing worse than simple rule-based models. To address this limitation, we propose a novel framework, Adaptive Prediction Ensemble (APE), which integrates deep learning and rule-based prediction experts. A learned routing function, trained concurrently with the deep learning model, dynamically selects the most reliable prediction based on the input scenario. Our experiments on large-scale datasets, including Waymo Open Motion Dataset (WOMD) and Argoverse, demonstrate improvement in zero-shot generalization across datasets. We show that our method outperforms individual prediction models and other variants, particularly in long-horizon prediction and scenarios with a high proportion of OOD data. This work highlights the potential of hybrid approaches for robust and generalizable motion prediction in autonomous driving. More details can be found on the project page: https://sites.google.com/view/ape-generalization.

Yifan Xue, Toktam Mohammadnejad, Faizan M Tariq et al.

Autonomous parking requires efficient path planning that ensures kinematic feasibility and collision avoidance in constrained environments. Hybrid A* is widely used but computationally expensive, while reinforcement learning (RL) methods lack reliability and often struggle with long-horizon geometric constraints, leading to suboptimal trajectories. We present N3P, a fast learning-based three-stage framework for automated parking. By introducing an intermediate preparatory pose and using a learning module to predict it, N3P decomposes the maneuver into simpler subproblems, thereby reducing computational complexity and accelerating path generation. We validate the framework by integrating it with Hybrid A* algorithms. Experiments in perpendicular and parallel parking scenarios show that N3P-enhanced Hybrid A* speeds up planning by more than 80%. It also outperforms RL baselines in success rate and trajectory quality, producing shorter trajectories with fewer gear changes, while achieving comparable or lower planning time in most cases.

Chak Lam Shek, Faizan M. Tariq, Sangjae Bae et al.

Heterogeneous multi-robot systems are increasingly deployed in long-horizon missions that require coordination among robots with diverse capabilities. However, existing planning approaches struggle to construct accurate symbolic representations and maintain plan consistency in dynamic environments. Classical PDDL planners require manually crafted symbolic models, while LLM-based planners often ignore agent heterogeneity and environmental uncertainty. We introduce KGLAMP, a knowledge-graph-guided LLM planning framework for heterogeneous multi-robot teams. The framework maintains a structured knowledge graph encoding object relations, spatial reachability, and robot capabilities, which guides the LLM in generating accurate PDDL problem specifications. The knowledge graph serves as a persistent, dynamically updated memory that incorporates new observations and triggers replanning upon detecting inconsistencies, enabling symbolic plans to adapt to evolving world states. Experiments on the MAT-THOR benchmark show that KGLAMP improves performance by at least 25.5% over both LLM-only and PDDL-based variants.

Piyush Gupta, David Isele, Enna Sachdeva et al.

We present a novel mission-planning strategy for heterogeneous multi-robot teams, taking into account the specific constraints and capabilities of each robot. Our approach employs hierarchical trees to systematically break down complex missions into manageable sub-tasks. We develop specialized APIs and tools, which are utilized by Large Language Models (LLMs) to efficiently construct these hierarchical trees. Once the hierarchical tree is generated, it is further decomposed to create optimized schedules for each robot, ensuring adherence to their individual constraints and capabilities. We demonstrate the effectiveness of our framework through detailed examples covering a wide range of missions, showcasing its flexibility and scalability.

Piyush Gupta, David Isele, Sangjae Bae

Real-world driving involves intricate interactions among vehicles navigating through dense traffic scenarios. Recent research focuses on enhancing the interaction awareness of autonomous vehicles to leverage these interactions in decision-making. These interaction-aware planners rely on neural-network-based prediction models to capture inter-vehicle interactions, aiming to integrate these predictions with traditional control techniques such as Model Predictive Control. However, this integration of deep learning-based models with traditional control paradigms often results in computationally demanding optimization problems, relying on heuristic methods. This study introduces a principled and efficient method for combining deep learning with constrained optimization, employing knowledge distillation to train smaller and more efficient networks, thereby mitigating complexity. We demonstrate that these refined networks maintain the problem-solving efficacy of larger models while significantly accelerating optimization. Specifically, in the domain of interaction-aware trajectory planning for autonomous vehicles, we illustrate that training a smaller prediction network using knowledge distillation speeds up optimization without sacrificing accuracy.

Anjian Li, Sangjae Bae, David Isele et al.

Trajectory prediction and planning are essential for autonomous vehicles to navigate safely and efficiently in dynamic environments. Traditional approaches often treat them separately, limiting the ability for interactive planning. While recent diffusion-based generative models have shown promise in multi-agent trajectory generation, their slow sampling is less suitable for high-frequency planning tasks. In this paper, we leverage the consistency model to build a predictive planner that samples from a joint distribution of ego and surrounding agents, conditioned on the ego vehicle's navigational goal. Trained on real-world human driving datasets, our consistency model generates higher-quality trajectories with fewer sampling steps than standard diffusion models, making it more suitable for real-time deployment. To enforce multiple planning constraints simultaneously on the ego trajectory, a novel online guided sampling approach inspired by the Alternating Direction Method of Multipliers (ADMM) is introduced. Evaluated on the Waymo Open Motion Dataset (WOMD), our method enables proactive behavior such as nudging and yielding, and also demonstrates smoother, safer, and more efficient trajectories and satisfaction of multiple constraints under a limited computational budget.

Longchao Da, David Isele, Hua Wei et al.

Being able to anticipate the motion of surrounding agents is essential for the safe operation of autonomous driving systems in dynamic situations. While various methods have been proposed for trajectory prediction, the current evaluation practices still rely on error-based metrics (e.g., ADE, FDE), which reveal the accuracy from a post-hoc view but ignore the actual effect the predictor brings to the self-driving vehicles (SDVs), especially in complex interactive scenarios: a high-quality predictor not only chases accuracy, but should also captures all possible directions a neighbor agent might move, to support the SDVs' cautious decision-making. Given that the existing metrics hardly account for this standard, in our work, we propose a comprehensive pipeline that adaptively evaluates the predictor's performance by two dimensions: accuracy and diversity. Based on the criticality of the driving scenario, these two dimensions are dynamically combined and result in a final score for the predictor's performance. Extensive experiments on a closed-loop benchmark using real-world datasets show that our pipeline yields a more reasonable evaluation than traditional metrics by better reflecting the correlation of the predictors' evaluation with the autonomous vehicles' driving performance. This evaluation pipeline shows a robust way to select a predictor that potentially contributes most to the SDV's driving performance.

Piyush Gupta, Sangjae Bae, David Isele

The adoption of Large Language Models (LLMs) is rapidly expanding across various tasks that involve inherent graphical structures. Graphs are integral to a wide range of applications, including motion planning for autonomous vehicles, social networks, scene understanding, and knowledge graphs. Many problems, even those not initially perceived as graph-based, can be effectively addressed through graph theory. However, when applied to these tasks, LLMs often encounter challenges, such as hallucinations and mathematical inaccuracies. To overcome these limitations, we propose Graph-Grounded LLMs, a system that improves LLM performance on graph-related tasks by integrating a graph library through function calls. By grounding LLMs in this manner, we demonstrate significant reductions in hallucinations and improved mathematical accuracy in solving graph-based problems, as evidenced by the performance on the NLGraph benchmark. Finally, we showcase a disaster rescue application where the Graph-Grounded LLM acts as a decision-support system.

Keuntaek Lee, David Isele, Evangelos A. Theodorou et al.

It can be difficult to autonomously produce driver behavior so that it appears natural to other traffic participants. Through Inverse Reinforcement Learning (IRL), we can automate this process by learning the underlying reward function from human demonstrations. We propose a new IRL algorithm that learns a goal-conditioned spatiotemporal reward function. The resulting costmap is used by Model Predictive Controllers (MPCs) to perform a task without any hand-designing or hand-tuning of the cost function. We evaluate our proposed Goal-conditioned SpatioTemporal Zeroing Maximum Entropy Deep IRL (GSTZ)-MEDIRL framework together with MPC in the CARLA simulator for autonomous driving, lane keeping, and lane changing tasks in a challenging dense traffic highway scenario. Our proposed methods show higher success rates compared to other baseline methods including behavior cloning, state-of-the-art RL policies, and MPC with a learning-based behavior prediction model.

Ran Tian, Liting Sun, Masayoshi Tomizuka et al.

A human-centered robot needs to reason about the cognitive limitation and potential irrationality of its human partner to achieve seamless interactions. This paper proposes an anytime game-theoretic planner that integrates iterative reasoning models, a partially observable Markov decision process, and chance-constrained Monte-Carlo belief tree search for robot behavioral planning. Our planner enables a robot to safely and actively reason about its human partner's latent cognitive states (bounded intelligence and irrationality) in real-time to maximize its utility better. We validate our approach in an autonomous driving domain where our behavioral planner and a low-level motion controller hierarchically control an autonomous car to negotiate traffic merges. Simulations and user studies are conducted to show our planner's effectiveness.

Sangjae Bae, David Isele, Kikuo Fujimura et al.

This paper proposes a discretionary lane selection algorithm. In particular, highway driving is considered as a targeted scenario, where each lane has a different level of traffic flow. When lane-changing is discretionary, it is advised not to change lanes unless highly beneficial, e.g., reducing travel time significantly or securing higher safety. Evaluating such "benefit" is a challenge, along with multiple surrounding vehicles in dynamic speed and heading with uncertainty. We propose a real-time lane-selection algorithm with careful cost considerations and with modularity in design. The algorithm is a search-based optimization method that evaluates uncertain dynamic positions of other vehicles under a continuous time and space domain. For demonstration, we incorporate a state-of-the-art motion planner framework (Neural Networks integrated Model Predictive Control) under a CARLA simulation environment.

Xiaobai Ma, Jiachen Li, Mykel J. Kochenderfer et al.

Deep reinforcement learning (DRL) provides a promising way for learning navigation in complex autonomous driving scenarios. However, identifying the subtle cues that can indicate drastically different outcomes remains an open problem with designing autonomous systems that operate in human environments. In this work, we show that explicitly inferring the latent state and encoding spatial-temporal relationships in a reinforcement learning framework can help address this difficulty. We encode prior knowledge on the latent states of other drivers through a framework that combines the reinforcement learner with a supervised learner. In addition, we model the influence passing between different vehicles through graph neural networks (GNNs). The proposed framework significantly improves performance in the context of navigating T-intersections compared with state-of-the-art baseline approaches.

Maxime Bouton, Alireza Nakhaei, David Isele et al.

Maneuvering in dense traffic is a challenging task for autonomous vehicles because it requires reasoning about the stochastic behaviors of many other participants. In addition, the agent must achieve the maneuver within a limited time and distance. In this work, we propose a combination of reinforcement learning and game theory to learn merging behaviors. We design a training curriculum for a reinforcement learning agent using the concept of level-$k$ behavior. This approach exposes the agent to a broad variety of behaviors during training, which promotes learning policies that are robust to model discrepancies. We show that our approach learns more efficient policies than traditional training methods.

Anahita Mohseni-Kabir, David Isele, Kikuo Fujimura

In a multi-agent setting, the optimal policy of a single agent is largely dependent on the behavior of other agents. We investigate the problem of multi-agent reinforcement learning, focusing on decentralized learning in non-stationary domains for mobile robot navigation. We identify a cause for the difficulty in training non-stationary policies: mutual adaptation to sub-optimal behaviors, and we use this to motivate a curriculum-based strategy for learning interactive policies. The curriculum has two stages. First, the agent leverages policy gradient algorithms to learn a policy that is capable of achieving multiple goals. Second, the agent learns a modifier policy to learn how to interact with other agents in a multi-agent setting. We evaluated our approach on both an autonomous driving lane-change domain and a robot navigation domain.

David Isele

Dense urban traffic environments can produce situations where accurate prediction and dynamic models are insufficient for successful autonomous vehicle motion planning. We investigate how an autonomous agent can safely negotiate with other traffic participants, enabling the agent to handle potential deadlocks. Specifically we consider merges where the gap between cars is smaller than the size of the ego vehicle. We propose a game theoretic framework capable of generating and responding to interactive behaviors. Our main contribution is to show how game-tree decision making can be executed by an autonomous vehicle, including approximations and reasoning that make the tree-search computationally tractable. Additionally, to test our model we develop a stochastic rule-based traffic agent capable of generating interactive behaviors that can be used as a benchmark for simulating traffic participants in a crowded merge setting.

David Isele, Alireza Nakhaei, Kikuo Fujimura

There have been numerous advances in reinforcement learning, but the typically unconstrained exploration of the learning process prevents the adoption of these methods in many safety critical applications. Recent work in safe reinforcement learning uses idealized models to achieve their guarantees, but these models do not easily accommodate the stochasticity or high-dimensionality of real world systems. We investigate how prediction provides a general and intuitive framework to constraint exploration, and show how it can be used to safely learn intersection handling behaviors on an autonomous vehicle.

Yuchen Cui, David Isele, Scott Niekum et al.

Estimating statistical uncertainties allows autonomous agents to communicate their confidence during task execution and is important for applications in safety-critical domains such as autonomous driving. In this work, we present the uncertainty-aware imitation learning (UAIL) algorithm for improving end-to-end control systems via data aggregation. UAIL applies Monte Carlo Dropout to estimate uncertainty in the control output of end-to-end systems, using states where it is uncertain to selectively acquire new training data. In contrast to prior data aggregation algorithms that force human experts to visit sub-optimal states at random, UAIL can anticipate its own mistakes and switch control to the expert in order to prevent visiting a series of sub-optimal states. Our experimental results from simulated driving tasks demonstrate that our proposed uncertainty estimation method can be leveraged to reliably predict infractions. Our analysis shows that UAIL outperforms existing data aggregation algorithms on a series of benchmark tasks.

Jiachen Yang, Alireza Nakhaei, David Isele et al.

A variety of cooperative multi-agent control problems require agents to achieve individual goals while contributing to collective success. This multi-goal multi-agent setting poses difficulties for recent algorithms, which primarily target settings with a single global reward, due to two new challenges: efficient exploration for learning both individual goal attainment and cooperation for others' success, and credit-assignment for interactions between actions and goals of different agents. To address both challenges, we restructure the problem into a novel two-stage curriculum, in which single-agent goal attainment is learned prior to learning multi-agent cooperation, and we derive a new multi-goal multi-agent policy gradient with a credit function for localized credit assignment. We use a function augmentation scheme to bridge value and policy functions across the curriculum. The complete architecture, called CM3, learns significantly faster than direct adaptations of existing algorithms on three challenging multi-goal multi-agent problems: cooperative navigation in difficult formations, negotiating multi-vehicle lane changes in the SUMO traffic simulator, and strategic cooperation in a Checkers environment.

David Isele, Akansel Cosgun

Deep reinforcement learning has emerged as a powerful tool for a variety of learning tasks, however deep nets typically exhibit forgetting when learning multiple tasks in sequence. To mitigate forgetting, we propose an experience replay process that augments the standard FIFO buffer and selectively stores experiences in a long-term memory. We explore four strategies for selecting which experiences will be stored: favoring surprise, favoring reward, matching the global training distribution, and maximizing coverage of the state space. We show that distribution matching successfully prevents catastrophic forgetting, and is consistently the best approach on all domains tested. While distribution matching has better and more consistent performance, we identify one case in which coverage maximization is beneficial - when tasks that receive less trained are more important. Overall, our results show that selective experience replay, when suitable selection algorithms are employed, can prevent catastrophic forgetting.

David Isele, Akansel Cosgun

We view intersection handling on autonomous vehicles as a reinforcement learning problem, and study its behavior in a transfer learning setting. We show that a network trained on one type of intersection generally is not able to generalize to other intersections. However, a network that is pre-trained on one intersection and fine-tuned on another performs better on the new task compared to training in isolation. This network also retains knowledge of the prior task, even though some forgetting occurs. Finally, we show that the benefits of fine-tuning hold when transferring simulated intersection handling knowledge to a real autonomous vehicle.

David Isele, Mohammad Rostami, Eric Eaton

Knowledge transfer between tasks can improve the performance of learned models, but requires an accurate estimate of the inter-task relationships to identify the relevant knowledge to transfer. These inter-task relationships are typically estimated based on training data for each task, which is inefficient in lifelong learning settings where the goal is to learn each consecutive task rapidly from as little data as possible. To reduce this burden, we develop a lifelong learning method based on coupled dictionary learning that utilizes high-level task descriptions to model the inter-task relationships. We show that using task descriptors improves the performance of the learned task policies, providing both theoretical justification for the benefit and empirical demonstration of the improvement across a variety of learning problems. Given only the descriptor for a new task, the lifelong learner is also able to accurately predict a model for the new task through zero-shot learning using the coupled dictionary, eliminating the need to gather training data before addressing the task.

David Isele, Akansel Cosgun, Kikuo Fujimura

We analyze how the knowledge to autonomously handle one type of intersection, represented as a Deep Q-Network, translates to other types of intersections (tasks). We view intersection handling as a deep reinforcement learning problem, which approximates the state action Q function as a deep neural network. Using a traffic simulator, we show that directly copying a network trained for one type of intersection to another type of intersection decreases the success rate. We also show that when a network that is pre-trained on Task A and then is fine-tuned on a Task B, the resulting network not only performs better on the Task B than an network exclusively trained on Task A, but also retained knowledge on the Task A. Finally, we examine a lifelong learning setting, where we train a single network on five different types of intersections sequentially and show that the resulting network exhibited catastrophic forgetting of knowledge on previous tasks. This result suggests a need for a long-term memory component to preserve knowledge.

David Isele, Reza Rahimi, Akansel Cosgun et al.

Providing an efficient strategy to navigate safely through unsignaled intersections is a difficult task that requires determining the intent of other drivers. We explore the effectiveness of Deep Reinforcement Learning to handle intersection problems. Using recent advances in Deep RL, we are able to learn policies that surpass the performance of a commonly-used heuristic approach in several metrics including task completion time and goal success rate and have limited ability to generalize. We then explore a system's ability to learn active sensing behaviors to enable navigating safely in the case of occlusions. Our analysis, provides insight into the intersection handling problem, the solutions learned by the network point out several shortcomings of current rule-based methods, and the failures of our current deep reinforcement learning system point to future research directions.