Agentic Robot: A Brain-Inspired Framework for Vision-Language-Action Models in Embodied Agents

2025-05-29

AI Summary

Long-horizon robotic manipulation faces critical challenges including error accumulation, absence of online execution verification, and insufficient robustness. To address these, we propose the brain-inspired Standardized Action Planning (SAP) framework—a closed-loop “plan–execute–verify” architecture: (1) a large language model decomposes high-level goals into executable sub-goals; (2) a vision-language-action model generates real-time, low-level control commands; and (3) a temporal sequence verifier performs introspective execution assessment. Inspired by human standard operating procedures (SOPs), SAP introduces the first dynamic self-verifying embodied multi-agent coordination framework. Evaluated on the LIBERO benchmark, SAP achieves a mean task success rate of 79.6%, surpassing SpatialVLA (+6.1%) and OpenVLA (+7.4%)—setting a new state-of-the-art. Moreover, SAP enhances system interpretability and reliability for real-world deployment.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Intelligent Robots: Manipulation
• Natural Language Processing: Safety and Robustness

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Long-horizon robotic manipulation poses significant challenges for autonomous systems, requiring extended reasoning, precise execution, and robust error recovery across complex sequential tasks. Current approaches, whether based on static planning or end-to-end visuomotor policies, suffer from error accumulation and lack effective verification mechanisms during execution, limiting their reliability in real-world scenarios. We present Agentic Robot, a brain-inspired framework that addresses these limitations through Standardized Action Procedures (SAP)--a novel coordination protocol governing component interactions throughout manipulation tasks. Drawing inspiration from Standardized Operating Procedures (SOPs) in human organizations, SAP establishes structured workflows for planning, execution, and verification phases. Our architecture comprises three specialized components: (1) a large reasoning model that decomposes high-level instructions into semantically coherent subgoals, (2) a vision-language-action executor that generates continuous control commands from real-time visual inputs, and (3) a temporal verifier that enables autonomous progression and error recovery through introspective assessment. This SAP-driven closed-loop design supports dynamic self-verification without external supervision. On the LIBERO benchmark, Agentic Robot achieves state-of-the-art performance with an average success rate of 79.6%, outperforming SpatialVLA by 6.1% and OpenVLA by 7.4% on long-horizon tasks. These results demonstrate that SAP-driven coordination between specialized components enhances both performance and interpretability in sequential manipulation, suggesting significant potential for reliable autonomous systems. Project Github: https://agentic-robot.github.io.

Problem

• Addresses error accumulation in robotic manipulation tasks
• Lacks effective verification mechanisms during execution
• Improves reliability in real-world sequential tasks

Innovation

• Brain-inspired framework for vision-language-action models
• Standardized Action Procedures for component coordination
• Specialized components for reasoning, execution, verification