Real-time Iteration Scheme for Diffusion Policy

Diffusion Policies have demonstrated impressive performance in robotic manipulation tasks. However, their long inference time, resulting from an extensive iterative denoising process, and the need to execute an action chunk before the next prediction to maintain consistent actions limit their applicability to latency-critical tasks or simple tasks with a short cycle time. While recent methods explored distillation or alternative policy structures to accelerate inference, these often demand additional training, which can be resource-intensive for large robotic models. In this paper, we introduce a novel approach inspired by the Real-Time Iteration (RTI) Scheme, a method from optimal control that accelerates optimization by leveraging solutions from previous time steps as initial guesses for subsequent iterations. We explore the application of this scheme in diffusion inference and propose a scaling-based method to effectively handle discrete actions, such as grasping, in robotic manipulation. The proposed scheme significantly reduces runtime computational costs without the need for distillation or policy redesign. This enables a seamless integration into many pre-trained diffusion-based models, in particular, to resource-demanding large models. We also provide theoretical conditions for the contractivity which could be useful for estimating the initial denoising step. Quantitative results from extensive simulation experiments show a substantial reduction in inference time, with comparable overall performance compared with Diffusion Policy using full-step denoising. Our project page with additional resources is available at: https://rti-dp.github.io/

Authors

• Yufei Duan
• Hang Yin
• Danica Kragic

Keywords

• Training
• Runtime
• Computational modeling
• Noise reduction
• Optimal control
• Real-time systems
• Iterative methods
• Low latency communication
• Optimization
• Intelligent robots
• Iterative Scheme
• Real-time Iteration
• Real-time Iteration Scheme
• Time Step
• Previous Step
• Simulation Experiments
• Inference Time
• Robot Manipulator
• Discrete Action
• Robot Model
• Robotic Tasks
• Shorter Cycle Time
• Horizon
• Bimodal
• Diffusion Process
• Sequence Of Actions
• Physical System
• Ordinary Differential Equations
• Reversible Process
• Nonlinear Model Predictive Control
• Diffusion Model
• Contractive
• Model Predictive Control
• Imitation Learning
• Fast Inference
• Stochastic Differential Equations
• Step Change
• Policy Learning
• Lipschitz Continuous