Robotics Research Hub
RoboRSI: Stable, efficient, and reusable robot self-evolution in complex real-world environments

RoboRSI: Stable, efficient, and reusable robot self-evolution in complex real-world environments

Zimo Wen, Yijin Chen, Yuxuan Cao, Wendi Chen, Yanwen Zou, Wenye Yu, Fuhang Kuang, Han Xue, Jun Lv, Chuan Wen, Cewu Lu · N/A · 2026

vision reinforcement-learning planning
PDF arXiv Code (GitHub)

Framework

N/A

License

N/A

Stars

N/A

Summary

A generalist robot should not only perform diverse tasks but also improve through experience, turning what it learns during execution into capabilities that later tasks can reuse.

Abstract Summary

A generalist robot should not only perform diverse tasks but also improve through experience, turning what it learns during execution into capabilities that later tasks can reuse. Robot agents that act through code can already repair programs from execution feedback, yet it remains a central challenge to organize this experience around the task structure that gives it meaning, so that each repair is attributed to the responsible capability, supported by execution evidence, and validated before it is reused. We introduce RoboRSI, a robot self-improvement system built on Top-Down Skill Refinement (TSR). TSR decomposes tasks into compound, atomic, and base skills with scoped responsibilities and explicit input--output contracts, attributes each execution outcome to the responsible branch, and confines revision to that branch. Building upon this structure, a Manager, Planner, Engineer, and Reviewer coordinate planning, execution, diagnosis, and the validated release of new skills, while people steer the process through objectives and corrections; stable skill sequences are further consolidated into reusable compound skills. On a mobile manipulator, RoboRSI develops multi-object household cleanup over 104 rounds. In simulation, it achieves the highest success rate on LIBERO, LIBERO-PRO, LIBERO-Plus, and RoboTwin, exceeding the strongest baseline by 2.7 to 11.0 percentage points.

Key Points

  • A generalist robot should not only perform diverse tasks but also improve through experience, turning what it learns during execution into capabilities that later tasks can reuse.
  • Robot agents that act through code can already repair programs from execution feedback, yet it remains a central challenge to organize this experience around the task structure that gives it meaning, so that each repair is attributed to the responsible capability, supported by execution evidence, and validated before it is reused.
  • We introduce RoboRSI, a robot self-improvement system built on Top-Down Skill Refinement (TSR)
  • TSR decomposes tasks into compound, atomic, and base skills with scoped responsibilities and explicit input--output contracts, attributes each execution outcome to the responsible branch, and confines revision to that branch.
  • Building upon this structure, a Manager, Planner, Engineer, and Reviewer coordinate planning, execution, diagnosis, and the validated release of new skills, while people steer the process through objectives and corrections; stable skill sequences are further consolidated into reusable compound skills.

Method

Consider a coding agent connected directly to robot tools and asked to improve a household cleanup task in a complex real-world home over many runs.

The context fills with records of earlier runs; a failure observed late in a sequence is patched where it appears, for example by adding retries to the release step, although it originated in perception or navigation; and because the objective is implicit in the latest prompt, successive edits optimize for the latest scene.

Method — RoboRSI: Stable, efficient, and reusable robot self-evolution in complex real-world environments
Figure 1: RoboRSI at a glance. The multi-agent self-improvement loop (center), real-world mobile manipulation and compound-skill gains (left), and simulation results (right).

Experiment

We report physical deployment and simulation results, then compare individual episodes and summarize the ablations. We study household floor cleanup with a WheelSingleArm M1 mobile manipulator. The robot searches for objects, approaches and grasps them, and transports them to designated receptacles. The case study covers 104 physical development runs with 24 hours of cumulative operation.

Before each run, the scene is restored to a common base layout, while the objects and receptacles are partially randomized; the navigation map remains fixed.

Experiment — RoboRSI: Stable, efficient, and reusable robot self-evolution in complex real-world environments
Figure 2: RoboRSI framework. Given a task, the Manager organizes its skill hierarchy, the Planner composes an execution plan, and the Engineer runs the required skills. The Reviewer diagnoses failures from observations and tool traces, returning them for TSR-guided revision. Tested updates enter the next iteration, and stable branches become parameterized skills for subsequent tasks.

Sources and Demonstrations

Figures are reproduced from Wen et al. See the full paper for experimental details and the project page, when the paper links one, for demonstrations and videos.

Share

Related Papers

AcousticDiffusion: Semantically Conditioned Audio-Guided Diffusion Policy for Search-and-Rescue Assistance
arXiv preprint

AcousticDiffusion: Semantically Conditioned Audio-Guided Diffusion Policy for Search-and-Rescue Assistance

Iana Zhura, Didar Seyidov, Dmitrii Plotnikov et al.

Navigating toward human callers is an important capability for rescue robots operating where visual contact is degraded or occluded. We present AcousticDiffusion, a semantically conditioned, audio-guided diffusion policy for human-directed navigation.

locomotion vision reinforcement-learning planning
PDF Advanced
Aim Short to Reach Far: Your Frozen World Model Can Plan Better Than You Think
arXiv preprint

Aim Short to Reach Far: Your Frozen World Model Can Plan Better Than You Think

Xvyuan Liu, Jianjie Fang, Chen Gao et al.

Planners built on visual world models commonly score each predicted outcome by its distance to the encoded goal image. We show that this target can limit control even with exact dynamics and globally optimal short-horizon search: reaching a goal may require actions that initially move away from it.

vision reinforcement-learning planning control human-robot-interaction benchmark
PDF Intermediate
Autonomous Droplet Navigation via Model-Based Reinforcement Learning: Zero-Shot Transfer and Emergent Dynamics
arXiv preprint

Autonomous Droplet Navigation via Model-Based Reinforcement Learning: Zero-Shot Transfer and Emergent Dynamics

Rajneesh Anand, Mayuresh V. Kothare

Self-driving laboratories (SDLs) are transforming chemical and materials discovery through closed-loop automation, yet automated infrastructure for physical manipulation of soft, deformable matter remains beyond current robotic platforms.

manipulation vision reinforcement-learning planning control
Code PDF Intermediate
GitHub ★ —