Gen-Drive: Enhancing Diffusion Generative Driving Policies with Reward Modeling and Reinforcement Learning Fine-tuning

Autonomous driving necessitates the ability to reason about future interactions between traffic agents and to make informed evaluations for planning. This paper introduces the \textit{Gen-Drive} framework, which shifts from the traditional prediction and deterministic planning framework to a generation-then-evaluation planning paradigm. The framework employs a behavior diffusion model as a scene generator to produce diverse possible future scenarios, thereby enhancing the capability for joint interaction reasoning. To facilitate decision-making, we propose a scene evaluator (reward) model, trained with pairwise preference data collected through VLM assistance, thereby reducing human workload and enhancing scalability. Furthermore, we utilize an RL fine-tuning framework to improve the generation quality of the diffusion model, rendering it more effective for planning tasks. We conduct training and closed-loop planning tests on the nuPlan dataset, and the results demonstrate that employing such a generation-then-evaluation strategy outperforms other learning-based approaches. Additionally, the fine-tuned generative driving policy shows significant enhancements in planning performance. We further demonstrate that utilizing our learned reward model for evaluation or RL fine-tuning leads to better planning performance compared to relying on human-designed rewards. Project website: https://mczhi.github.io/GenDrive.

NAVSIM: Data-Driven Non-Reactive Autonomous Vehicle Simulation and Benchmarking

Benchmarking vision-based driving policies is challenging. On one hand, open-loop evaluation with real data is easy, but these results do not reflect closed-loop performance. On the other, closed-loop evaluation is possible in simulation, but is hard to scale due to its significant computational demands. Further, the simulators available today exhibit a large domain gap to real data. This has resulted in an inability to draw clear conclusions from the rapidly growing body of research on end-to-end autonomous driving. In this paper, we present NAVSIM, a middle ground between these evaluation paradigms, where we use large datasets in combination with a non-reactive simulator to enable large-scale real-world benchmarking. Specifically, we gather simulation-based metrics, such as progress and time to collision, by unrolling bird's eye view abstractions of the test scenes for a short simulation horizon. Our simulation is non-reactive, i.e., the evaluated policy and environment do not influence each other. As we demonstrate empirically, this decoupling allows open-loop metric computation while being better aligned with closed-loop evaluations than traditional displacement errors. NAVSIM enabled a new competition held at CVPR 2024, where 143 teams submitted 463 entries, resulting in several new insights. On a large set of challenging scenarios, we observe that simple methods with moderate compute requirements such as TransFuser can match recent large-scale end-to-end driving architectures such as UniAD. Our modular framework can potentially be extended with new datasets, data curation strategies, and metrics, and will be continually maintained to host future challenges. Our code is available at https://github.com/autonomousvision/navsim.

Versatile Scene-Consistent Traffic Scenario Generation as Optimization with Diffusion

Generating realistic and controllable agent behaviors in traffic simulation is crucial for the development of autonomous vehicles. This problem is often formulated as imitation learning (IL) from real-world driving data by either directly predicting future trajectories or inferring cost functions with inverse optimal control. In this paper, we draw a conceptual connection between IL and diffusion-based generative modeling and introduce a novel framework Versatile Behavior Diffusion (VBD) to simulate interactive scenarios with multiple traffic participants. Our model not only generates scene-consistent multi-agent interactions but also enables scenario editing through multi-step guidance and refinement. Experimental evaluations show that VBD achieves state-of-the-art performance on the Waymo Sim Agents benchmark. In addition, we illustrate the versatility of our model by adapting it to various applications. VBD is capable of producing scenarios conditioning on priors, integrating with model-based optimization, sampling multi-modal scene-consistent scenarios by fusing marginal predictions, and generating safety-critical scenarios when combined with a game-theoretic solver.

Hybrid-Prediction Integrated Planning for Autonomous Driving

Autonomous driving systems require the ability to fully understand and predict the surrounding environment to make informed decisions in complex scenarios. Recent advancements in learning-based systems have highlighted the importance of integrating prediction and planning modules. However, this integration has brought forth three major challenges: inherent trade-offs by sole prediction, consistency between prediction patterns, and social coherence in prediction and planning. To address these challenges, we introduce a hybrid-prediction integrated planning (HPP) system, which possesses three novelly designed modules. First, we introduce marginal-conditioned occupancy prediction to align joint occupancy with agent-wise perceptions. Our proposed MS-OccFormer module achieves multi-stage alignment per occupancy forecasting with consistent awareness from agent-wise motion predictions. Second, we propose a game-theoretic motion predictor, GTFormer, to model the interactive future among individual agents with their joint predictive awareness. Third, hybrid prediction patterns are concurrently integrated with Ego Planner and optimized by prediction guidance. HPP achieves state-of-the-art performance on the nuScenes dataset, demonstrating superior accuracy and consistency for end-to-end paradigms in prediction and planning. Moreover, we test the long-term open-loop and closed-loop performance of HPP on the Waymo Open Motion Dataset and CARLA benchmark, surpassing other integrated prediction and planning pipelines with enhanced accuracy and compatibility.

Learning Online Belief Prediction for Efficient POMDP Planning in Autonomous Driving

Effective decision-making in autonomous driving relies on accurate inference of other traffic agents' future behaviors. To achieve this, we propose an online learning-based behavior prediction model and an efficient planner for Partially Observable Markov Decision Processes (POMDPs). We develop a learning-based prediction model, enhanced with a recurrent neural memory network, to dynamically update latent belief states and infer the intentions of other agents. The model can also integrate the ego vehicle's intentions to reflect closed-loop interactions among agents, and it learns from both offline data and online interactions. For planning, we employ an option-based Monte-Carlo Tree Search (MCTS) planner, which reduces computational complexity by searching over action sequences. Inside the MCTS planner, we use predicted long-term multi-modal trajectories to approximate future updates, which eliminates iterative belief updating and improves the running efficiency. Our approach also incorporates deep Q-learning (DQN) as a search prior, which significantly improves the performance of the MCTS planner. Experimental results from simulated environments validate the effectiveness of our proposed method. The online belief update model can significantly enhance the accuracy and temporal consistency of predictions, leading to improved decision-making performance. Employing DQN as a search prior in the MCTS planner considerably boosts its performance and outperforms an imitation learning-based prior. Additionally, we show that the option-based MCTS substantially outperforms the vanilla method in terms of performance and efficiency.

DTPP: Differentiable Joint Conditional Prediction and Cost Evaluation for Tree Policy Planning in Autonomous Driving

Motion prediction and cost evaluation are vital components in the decision-making system of autonomous vehicles. However, existing methods often ignore the importance of cost learning and treat them as separate modules. In this study, we employ a tree-structured policy planner and propose a differentiable joint training framework for both ego-conditioned prediction and cost models, resulting in a direct improvement of the final planning performance. For conditional prediction, we introduce a query-centric Transformer model that performs efficient ego-conditioned motion prediction. For planning cost, we propose a learnable context-aware cost function with latent interaction features, facilitating differentiable joint learning. We validate our proposed approach using the real-world nuPlan dataset and its associated planning test platform. Our framework not only matches state-of-the-art planning methods but outperforms other learning-based methods in planning quality, while operating more efficiently in terms of runtime. We show that joint training delivers significantly better performance than separate training of the two modules. Additionally, we find that tree-structured policy planning outperforms the conventional single-stage planning approach.

Occupancy Prediction-Guided Neural Planner for Autonomous Driving

Forecasting the scalable future states of surrounding traffic participants in complex traffic scenarios is a critical capability for autonomous vehicles, as it enables safe and feasible decision-making. Recent successes in learning-based prediction and planning have introduced two primary challenges: generating accurate joint predictions for the environment and integrating prediction guidance for planning purposes. To address these challenges, we propose a two-stage integrated neural planning framework, termed OPGP, that incorporates joint prediction guidance from occupancy forecasting. The preliminary planning phase simultaneously outputs the predicted occupancy for various types of traffic actors based on imitation learning objectives, taking into account shared interactions, scene context, and actor dynamics within a unified Transformer structure. Subsequently, the transformed occupancy prediction guides optimization to further inform safe and smooth planning under Frenet coordinates. We train our planner using a large-scale, real-world driving dataset and validate it in open-loop configurations. Our proposed planner outperforms strong learning-based methods, exhibiting improved performance due to occupancy prediction guidance.

GameFormer: Game-theoretic Modeling and Learning of Transformer-based Interactive Prediction and Planning for Autonomous Driving

Autonomous vehicles operating in complex real-world environments require accurate predictions of interactive behaviors between traffic participants. While existing works focus on modeling agent interactions based on their past trajectories, their future interactions are often ignored. This paper addresses the interaction prediction problem by formulating it with hierarchical game theory and proposing the GameFormer framework to implement it. Specifically, we present a novel Transformer decoder structure that uses the prediction results from the previous level together with the common environment background to iteratively refine the interaction process. Moreover, we propose a learning process that regulates an agent's behavior at the current level to respond to other agents' behaviors from the last level. Through experiments on a large-scale real-world driving dataset, we demonstrate that our model can achieve state-of-the-art prediction accuracy on the interaction prediction task. We also validate the model's capability to jointly reason about the ego agent's motion plans and other agents' behaviors in both open-loop and closed-loop planning tests, outperforming a variety of baseline methods.

Learning Interaction-aware Motion Prediction Model for Decision-making in Autonomous Driving

Predicting the behaviors of other road users is crucial to safe and intelligent decision-making for autonomous vehicles (AVs). However, most motion prediction models ignore the influence of the AV's actions and the planning module has to treat other agents as unalterable moving obstacles. To address this problem, this paper proposes an interaction-aware motion prediction model that is able to predict other agents' future trajectories according to the ego agent's future plan, i.e., their reactions to the ego's actions. Specifically, we employ Transformers to effectively encode the driving scene and incorporate the AV's plan in decoding the predicted trajectories. To train the model to accurately predict the reactions of other agents, we develop an online learning framework, where the ego agent explores the environment and collects other agents' reactions to itself. We validate the decision-making and learning framework in three highly interactive simulated driving scenarios. The results reveal that our decision-making method significantly outperforms the reinforcement learning methods in terms of data efficiency and performance. We also find that using the interaction-aware model can bring better performance than the non-interaction-aware model and the exploration process helps improve the success rate in testing.

Conditional Predictive Behavior Planning with Inverse Reinforcement Learning for Human-like Autonomous Driving

Making safe and human-like decisions is an essential capability of autonomous driving systems and learning-based behavior planning is a promising pathway toward this objective. Distinguished from existing learning-based methods that directly output decisions, this work introduces a predictive behavior planning framework that learns to predict and evaluate from human driving data. Concretely, a behavior generation module first produces a diverse set of candidate behaviors in the form of trajectory proposals. Then the proposed conditional motion prediction network is employed to forecast other agents' future trajectories conditioned on each trajectory proposal. Given the candidate plans and associated prediction results, we learn a scoring module to evaluate the plans using maximum entropy inverse reinforcement learning (IRL). We conduct comprehensive experiments to validate the proposed framework on a large-scale real-world urban driving dataset. The results reveal that the conditional prediction model is able to forecast multiple possible future trajectories given a candidate behavior and the prediction results are reactive to different plans. Moreover, the IRL-based scoring module can properly evaluate the trajectory proposals and select close-to-human ones. The proposed framework outperforms other baseline methods in terms of similarity to human driving trajectories. Moreover, we find that the conditional prediction model can improve both prediction and planning performance compared to the non-conditional model, and learning the scoring module is critical to correctly evaluating the candidate plans to align with human drivers.