Implicit Maximum Likelihood Estimation for Real-time Generative Model Predictive Control

2026-03-13

AI Summary

Diffusion models suffer from slow inference speeds in trajectory planning, hindering their applicability to real-time model predictive control (MPC) that demands rapid and continuous adaptation to dynamic environments. To address this limitation, this work proposes integrating Implicit Maximum Likelihood Estimation (IMLE) into generative trajectory planning, substantially accelerating inference while preserving multimodal trajectory generation capabilities. Evaluated on standard offline reinforcement learning benchmarks, the proposed method matches the performance of diffusion models in generating diverse trajectories yet achieves planning speeds two orders of magnitude faster in both open-loop and closed-loop settings. The approach further demonstrates effective real-time adaptive planning in human-robot navigation scenarios, confirming its suitability for latency-sensitive MPC applications.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Computer Vision: Diffusion Models for Vision

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Diffusion-based models have recently shown strong performance in trajectory planning, as they are capable of capturing diverse, multimodal distributions of complex behaviors. A key limitation of these models is their slow inference speed, which results from the iterative denoising process. This makes them less suitable for real-time applications such as closed-loop model predictive control (MPC), where plans must be generated quickly and adapted continuously to a changing environment. In this paper, we investigate Implicit Maximum Likelihood Estimation (IMLE) as an alternative generative modeling approach for planning. IMLE offers strong mode coverage while enabling inference that is two orders of magnitude faster, making it particularly well suited for real-time MPC tasks. Our results demonstrate that IMLE achieves competitive performance on standard offline reinforcement learning benchmarks compared to the standard diffusion-based planner, while substantially improving planning speed in both open-loop and closed-loop settings. We further validate IMLE in a closed-loop human navigation scenario, operating in real-time, demonstrating how it enables rapid and adaptive plan generation in dynamic environments.

Problem

• diffusion models
• real-time planning
• model predictive control
• inference speed
• trajectory planning

Innovation

• Implicit Maximum Likelihood Estimation
• Real-time MPC
• Generative Planning
• Fast Inference
• Trajectory Optimization