FrankenBot: Brain-Morphic Modular Orchestration for Robotic Manipulation with Vision-Language Models

TL;DR

FrankenBot is a brain-inspired, modular robotic manipulation framework leveraging vision-language models to integrate multiple cognitive functions, achieving high efficiency and robustness in complex environments without retraining.

Contribution

The paper introduces FrankenBot, a novel VLM-driven, brain-morphic architecture that unifies key robotic functions into a cohesive system inspired by human brain structure, enhancing efficiency and capability.

Findings

Improves anomaly detection and handling in robotic manipulation.

Enables long-term memory and stable operation without retraining.

Demonstrates superior performance in simulation and real-world tests.

Abstract

Developing a general robot manipulation system capable of performing a wide range of tasks in complex, dynamic, and unstructured real-world environments has long been a challenging task. It is widely recognized that achieving human-like efficiency and robustness manipulation requires the robotic brain to integrate a comprehensive set of functions, such as task planning, policy generation, anomaly monitoring and handling, and long-term memory, achieving high-efficiency operation across all functions. Vision-Language Models (VLMs), pretrained on massive multimodal data, have acquired rich world knowledge, exhibiting exceptional scene understanding and multimodal reasoning capabilities. However, existing methods typically focus on realizing only a single function or a subset of functions within the robotic brain, without integrating them into a unified cognitive architecture. Inspired by a divide-and-conquer strategy and the architecture of the human brain, we propose FrankenBot, a VLM-driven, brain-morphic robotic manipulation framework that achieves both comprehensive functionality and high operational efficiency. Our framework includes a suite of components, decoupling a part of key functions from frequent VLM calls, striking an optimal balance between functional completeness and system efficiency. Specifically, we map task planning, policy generation, memory management, and low-level interfacing to the cortex, cerebellum, temporal lobe-hippocampus complex, and brainstem, respectively, and design efficient coordination mechanisms for the modules. We conducted comprehensive experiments in both simulation and real-world robotic environments, demonstrating that our method offers significant advantages in anomaly detection and handling, long-term memory, operational efficiency, and stability -- all without requiring any fine-tuning or retraining.