Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

LiMoDE: Rethinking Lifelong Robot Manipulation from a Mixture-of-Dynamic-Experts Perspective

About

Building a generalist robot that can leverage prior knowledge for continuous task adaptation remains a significant challenge. Previous works alleviate the catastrophic forgetting problem by parameter-efficient fine-tuning for single-task adaptation. However, they fail to extract reusable skills and model the interaction with other skills effectively. Recent works try to address these issues by learning prompts. Differently, this paper presents an architectural perspective on the Lifelong Mixture of Dynamic Experts (\textit{LiMoDE}), a novel two-stage learning scheme for lifelong robot manipulation. Specifically, a dynamic MoE structure is first proposed in the multi-task pre-training stage to learn prior knowledge, where a varied number of heterogeneous experts are activated based on the motion information to address different short-term manipulations. Subsequently, in the task adaptation stage, we design a lifelong MoE adaptation mechanism % (LiMoEAM) that learns lifelong experts and dynamically combines them with frozen ones for new tasks, facilitating the knowledge transfer during adaptation. The proposed \textit{LiMoDE} is evaluated on both the simulated lifelong learning benchmark and real-world tasks. Extensive experiments demonstrate its effectiveness in achieving superior performance and strong lifelong adaptation by introducing a moderate number of additional trainable parameters and inference overhead.

Zhihao Gu, Lin Wang• 2026

Related benchmarks

TaskDatasetResultRank
Continual LearningLIBERO Object
FWT81
25
Multi-task LearningLIBERO
Object Score94.2
25
Robotic ManipulationReal-world Lifelong Tasks
Success Rate65
15
Continual LearningLIBERO Long
Forward Transfer (FWT)55
15
Continual LearningLIBERO Spatial
FWT75
14
Lifelong LearningLIBERO Goal
Forward Transfer (FWT)70
10
Robotic ManipulationReal-world Pre-training Tasks
Success Rate70
9
Lifelong robotic manipulationLIBERO Lifelong Learning (FWT)
OBJECT Success Rate81
4
Showing 8 of 8 rows

Other info

Follow for update