Diffusion Modulation via Environment Mechanism Modeling for Planning
Diffusion models have shown promising capabilities in trajectory generation for planning in offline reinforcement learning (RL). However, conventional diffusion-based planning methods often fail to account for the fact that generating trajectories in RL requires unique consistency between transitions to ensure coherence in real environments. This oversight can result in considerable discrepancies between the generated trajectories and the underlying mechanisms of a real environment. To address this problem, we propose a novel diffusion-based planning method, termed as Diffusion Modulation via Environment Mechanism Modeling (DMEMM). DMEMM modulates diffusion model training by incorporating key RL environment mechanisms, particularly transition dynamics and reward functions. Experimental results demonstrate that DMEMM achieves state-of-the-art performance for planning with offline reinforcement learning.
Code (0)
등록된 구현이 없습니다.
Tasks
Reinforcement LearningSimilar Papers 제목 키워드 기반
Diffusion-Based Planning for Autonomous Driving with Flexible Guidance
Achieving human-like driving behaviors in complex open-world environments is a critical challenge in autonomous driving. Contemporary learning-based planning approaches such as imitation learning methods often struggle t…
Autonomous DrivingImitation LearningValdi: Value Diffusion World Models
World models can enable Model Predictive Control (MPC), but this requires dynamics prediction that is both fast enough for online use and expressive enough to represent uncertain futures. Diffusion models offer a natural…
Constrained Generative Modeling with Manually Bridged Diffusion Models
In this paper we describe a novel framework for diffusion-based generative modeling on constrained spaces. In particular, we introduce manual bridges, a framework that expands the kinds of constraints that can be practic…
Autonomous VehiclesA Diffusion-Refined Planner with Reinforcement Learning Priors for Confined-Space Parking
The growing demand for parking has increased the need for automated parking planning methods that can operate reliably in confined spaces. In restricted and complex environments, high-precision maneuvers are required to …
Reinforcement LearningSCORP: Scene-Consistent Multi-agent Diffusion Planning with Stable Online Reinforcement Post-Training for Cooperative Driving
Cooperative driving is a safety- and efficiency-critical task that requires the coordination of diverse, interaction-realistic multi-agent trajectories. Although existing diffusion-based methods can capture multimodal be…
Reinforcement Learning