WiseAD: Knowledge Augmented End-to-End Autonomous Driving with Vision-Language Model

The emergence of general human knowledge and impressive logical reasoning capacity in rapidly progressed vision-language models (VLMs) have driven increasing interest in applying VLMs to high-level autonomous driving tasks, such as scene understanding and decision-making. However, an in-depth study on the relationship between knowledge proficiency, especially essential driving expertise, and closed-loop autonomous driving performance requires further exploration. In this paper, we investigate the effects of the depth and breadth of fundamental driving knowledge on closed-loop trajectory planning and introduce WiseAD, a specialized VLM tailored for end-to-end autonomous driving capable of driving reasoning, action justification, object recognition, risk analysis, driving suggestions, and trajectory planning across diverse scenarios. We employ joint training on driving knowledge and planning datasets, enabling the model to perform knowledge-aligned trajectory planning accordingly. Extensive experiments indicate that as the diversity of driving knowledge extends, critical accidents are notably reduced, contributing 11.9% and 12.4% improvements in the driving score and route completion on the Carla closed-loop evaluations, achieving state-of-the-art performance. Moreover, WiseAD also demonstrates remarkable performance in knowledge evaluations on both in-domain and out-of-domain datasets.

Integrating Decision-Making Into Differentiable Optimization Guided Learning for End-to-End Planning of Autonomous Vehicles

We address the decision-making capability within an end-to-end planning framework that focuses on motion prediction, decision-making, and trajectory planning. Specifically, we formulate decision-making and trajectory planning as a differentiable nonlinear optimization problem, which ensures compatibility with learning-based modules to establish an end-to-end trainable architecture. This optimization introduces explicit objectives related to safety, traveling efficiency, and riding comfort, guiding the learning process in our proposed pipeline. Intrinsic constraints resulting from the decision-making task are integrated into the optimization formulation and preserved throughout the learning process. By integrating the differentiable optimizer with a neural network predictor, the proposed framework is end-to-end trainable, aligning various driving tasks with ultimate performance goals defined by the optimization objectives. The proposed framework is trained and validated using the Waymo Open Motion dataset. The open-loop testing reveals that while the planning outcomes using our method do not always resemble the expert trajectory, they consistently outperform baseline approaches with improved safety, traveling efficiency, and riding comfort. The closed-loop testing further demonstrates the effectiveness of optimizing decisions and improving driving performance. Ablation studies demonstrate that the initialization provided by the learning-based prediction module is essential for the convergence of the optimizer as well as the overall driving performance.

V2X-Radar: A Multi-modal Dataset with 4D Radar for Cooperative Perception

Modern autonomous vehicle perception systems often struggle with occlusions and limited perception range. Previous studies have demonstrated the effectiveness of cooperative perception in extending the perception range and overcoming occlusions, thereby improving the safety of autonomous driving. In recent years, a series of cooperative perception datasets have emerged. However, these datasets only focus on camera and LiDAR, overlooking 4D Radar, a sensor employed in single-vehicle autonomous driving for robust perception in adverse weather conditions. In this paper, to bridge the gap of missing 4D Radar datasets in cooperative perception, we present V2X-Radar, the first large real-world multi-modal dataset featuring 4D Radar. Our V2X-Radar dataset is collected using a connected vehicle platform and an intelligent roadside unit equipped with 4D Radar, LiDAR, and multi-view cameras. The collected data includes sunny and rainy weather conditions, spanning daytime, dusk, and nighttime, as well as typical challenging scenarios. The dataset comprises 20K LiDAR frames, 40K camera images, and 20K 4D Radar data, with 350K annotated bounding boxes across five categories. To facilitate diverse research domains, we establish V2X-Radar-C for cooperative perception, V2X-Radar-I for roadside perception, and V2X-Radar-V for single-vehicle perception. We further provide comprehensive benchmarks of recent perception algorithms on the above three sub-datasets. The dataset and benchmark codebase will be available at \url{http://openmpd.com/column/V2X-Radar}.

Path Planning in Complex Environments with Superquadrics and Voronoi-Based Orientation

Path planning in narrow passages is a challenging problem in various applications. Traditional planning algorithms often face challenges in complex environments like mazes and traps, where narrow entrances require special orientation control for successful navigation. In this work, we present a novel approach that combines superquadrics (SQ) representation and Voronoi diagrams to solve the narrow passage problem in both 2D and 3D environment. Our method utilizes the SQ formulation to expand obstacles, eliminating impassable passages, while Voronoi hyperplane ensures maximum clearance path. Additionally, the hyperplane provides a natural reference for robot orientation, aligning its long axis with the passage direction. We validate our framework through a 2D object retrieval task and 3D drone simulation, demonstrating that our approach outperforms classical planners and a cutting-edge drone planner by ensuring passable trajectories with maximum clearance.

Planning for quasi-static manipulation tasks via an intrinsic haptic metric

Contact-rich manipulation often requires strategic interactions with objects, such as pushing to accomplish specific tasks. We propose a novel scenario where a robot inserts a book into a crowded shelf by pushing aside neighboring books to create space before slotting the new book into place. Classical planning algorithms fail in this context due to limited space and their tendency to avoid contact. Additionally, they do not handle indirectly manipulable objects or consider force interactions. Our key contributions are: i) re-framing quasi-static manipulation as a planning problem on an implicit manifold derived from equilibrium conditions; ii) utilizing an intrinsic haptic metric instead of ad-hoc cost functions; and iii) proposing an adaptive algorithm that simultaneously updates robot states, object positions, contact points, and haptic distances. We evaluate our method on such crowded bookshelf insertion task but it is a general formulation to rigid bodies manipulation tasks. We propose proxies to capture contact point and force, with superellipse to represent objects. This simplified model guarantee the differentiablity. Our framework autonomously discovers strategic wedging-in policies while our simplified contact model achieves behavior similar to real world scenarios. We also vary the stiffness and initial positions to analysis our framework comprehensively. The video can be found at https://youtu.be/eab8umZ3AQ0.

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.

Reasoning Multi-Agent Behavioral Topology for Interactive Autonomous Driving

Autonomous driving system aims for safe and social-consistent driving through the behavioral integration among interactive agents. However, challenges remain due to multi-agent scene uncertainty and heterogeneous interaction. Current dense and sparse behavioral representations struggle with inefficiency and inconsistency in multi-agent modeling, leading to instability of collective behavioral patterns when integrating prediction and planning (IPP). To address this, we initiate a topological formation that serves as a compliant behavioral foreground to guide downstream trajectory generations. Specifically, we introduce Behavioral Topology (BeTop), a pivotal topological formulation that explicitly represents the consensual behavioral pattern among multi-agent future. BeTop is derived from braid theory to distill compliant interactive topology from multi-agent future trajectories. A synergistic learning framework (BeTopNet) supervised by BeTop facilitates the consistency of behavior prediction and planning within the predicted topology priors. Through imitative contingency learning, BeTop also effectively manages behavioral uncertainty for prediction and planning. Extensive verification on large-scale real-world datasets, including nuPlan and WOMD, demonstrates that BeTop achieves state-of-the-art performance in both prediction and planning tasks. Further validations on the proposed interactive scenario benchmark showcase planning compliance in interactive cases.

Risk-Aware Vehicle Trajectory Prediction Under Safety-Critical Scenarios

Trajectory prediction is significant for intelligent vehicles to achieve high-level autonomous driving, and a lot of relevant research achievements have been made recently. Despite the rapid development, most existing studies solely focused on normal safe scenarios while largely neglecting safety-critical scenarios, particularly those involving imminent collisions. This oversight may result in autonomous vehicles lacking the essential predictive ability in such situations, posing a significant threat to safety. To tackle these, this paper proposes a risk-aware trajectory prediction framework tailored to safety-critical scenarios. Leveraging distinctive hazardous features, we develop three core risk-aware components. First, we introduce a risk-incorporated scene encoder, which augments conventional encoders with quantitative risk information to achieve risk-aware encoding of hazardous scene contexts. Next, we incorporate endpoint-risk-combined intention queries as prediction priors in the decoder to ensure that the predicted multimodal trajectories cover both various spatial intentions and risk levels. Lastly, an auxiliary risk prediction task is implemented for the ultimate risk-aware prediction. Furthermore, to support model training and performance evaluation, we introduce a safety-critical trajectory prediction dataset and tailored evaluation metrics. We conduct comprehensive evaluations and compare our model with several SOTA models. Results demonstrate the superior performance of our model, with a significant improvement in most metrics. This prediction advancement enables autonomous vehicles to execute correct collision avoidance maneuvers under safety-critical scenarios, eventually enhancing road traffic safety.

Variational Delayed Policy Optimization

In environments with delayed observation, state augmentation by including actions within the delay window is adopted to retrieve Markovian property to enable reinforcement learning (RL). However, state-of-the-art (SOTA) RL techniques with Temporal-Difference (TD) learning frameworks often suffer from learning inefficiency, due to the significant expansion of the augmented state space with the delay. To improve learning efficiency without sacrificing performance, this work introduces a novel framework called Variational Delayed Policy Optimization (VDPO), which reformulates delayed RL as a variational inference problem. This problem is further modelled as a two-step iterative optimization problem, where the first step is TD learning in the delay-free environment with a small state space, and the second step is behaviour cloning which can be addressed much more efficiently than TD learning. We not only provide a theoretical analysis of VDPO in terms of sample complexity and performance, but also empirically demonstrate that VDPO can achieve consistent performance with SOTA methods, with a significant enhancement of sample efficiency (approximately 50\% less amount of samples) in the MuJoCo benchmark.

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.