HyCodePolicy: Hybrid Language Controllers for Multimodal Monitoring and Decision in Embodied Agents

TL;DR

HyCodePolicy introduces a hybrid control framework for embodied agents that combines code generation, geometric grounding, perceptual monitoring, and iterative repair to enhance robustness and efficiency in task execution.

Contribution

It presents a novel hybrid language-based control system that integrates multiple modalities and feedback mechanisms for adaptive policy repair in embodied agents.

Findings

Significantly improves robustness of robot manipulation policies.

Enhances sample efficiency in policy learning.

Enables self-correcting program synthesis with minimal supervision.

Abstract

Recent advances in multimodal large language models (MLLMs) have enabled richer perceptual grounding for code policy generation in embodied agents. However, most existing systems lack effective mechanisms to adaptively monitor policy execution and repair codes during task completion. In this work, we introduce HyCodePolicy, a hybrid language-based control framework that systematically integrates code synthesis, geometric grounding, perceptual monitoring, and iterative repair into a closed-loop programming cycle for embodied agents. Technically, given a natural language instruction, our system first decomposes it into subgoals and generates an initial executable program grounded in object-centric geometric primitives. The program is then executed in simulation, while a vision-language model (VLM) observes selected checkpoints to detect and localize execution failures and infer failure reasons. By fusing structured execution traces capturing program-level events with VLM-based perceptual feedback, HyCodePolicy infers failure causes and repairs programs. This hybrid dual feedback mechanism enables self-correcting program synthesis with minimal human supervision. Our results demonstrate that HyCodePolicy significantly improves the robustness and sample efficiency of robot manipulation policies, offering a scalable strategy for integrating multimodal reasoning into autonomous decision-making pipelines.