Diverse Controllable Diffusion Policy With Signal Temporal Logic

2024-10-01

IEEE Robotics and Automation Letters

AI Summary

To address poor controllability, limited diversity, and weak rule compliance of road agents in autonomous driving simulation, this paper introduces the first diffusion-based generative framework incorporating Signal Temporal Logic (STL) constraints. We propose a three-stage paradigm: STL calibration, trajectory optimization synthesis, and diffusion policy distillation. Our method explicitly encodes traffic rules into the generation process to ensure formal verifiability, enables STL-parameter-driven controllable trajectory synthesis, and simultaneously achieves high diversity and computational efficiency. Evaluated on nuScenes, our generated trajectories achieve 100% STL specification satisfaction and state-of-the-art diversity. Inference speed is 17× faster than the second-best method. In closed-loop simulation, our approach attains the lowest collision rate, along with significantly higher rule compliance and behavioral diversity compared to existing methods.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Planning, Routing, and Scheduling: Planning with Language Models
• Multiagent Systems: Agent-Based Simulation and Emergent Behavior

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Generating realistic simulations is critical for autonomous system applications such as self-driving and human-robot interactions. However, driving simulators nowadays still have difficulty in generating controllable, diverse, and rule-compliant behaviors for road participants: Rule-based models cannot produce diverse behaviors and require careful tuning, whereas learning-based methods imitate the policy from data but are not designed to follow the rules explicitly. Besides, the real-world datasets are by nature “single-outcome”, making the learning method hard to generate diverse behaviors. In this letter, we leverage Signal Temporal Logic (STL) and Diffusion Models to learn controllable, diverse, and rule-aware policy. We first calibrate the STL on the real-world data, then generate diverse synthetic data using trajectory optimization, and finally learn the rectified diffusion policy on the augmented dataset. We test on the NuScenes dataset and our approach can achieve the most diverse rule-compliant trajectories compared to other baselines, with a runtime 1/17X to the second-best approach. In the closed-loop testing, our approach reaches the highest diversity, rule satisfaction rate, and the least collision rate. Our method can generate varied characteristics conditional on different STL parameters in testing. A case study on human-robot encounter scenarios shows our approach can generate diverse and closed-to-oracle trajectories.

Problem

• Generating diverse, rule-compliant behaviors for autonomous systems.
• Overcoming limitations of rule-based and learning-based methods in simulations.
• Enhancing simulation realism with controllable and diverse trajectory generation.

Innovation

• Signal Temporal Logic for rule compliance
• Diffusion Models for diverse behavior generation
• Trajectory optimization for synthetic data creation