Search arXiv⌕ Search
Home/Archive/arXiv

arXiv · 2609.27612

RegenHarness: A Robot Agent Harness with Evidence-Gated Recursive Self-Improvement

Kailin Wang·Haoxiang Jie·Yaoyuan Yan·Zhiyou Heng·Zhaosong Li
View original source ↗Read source document ↗Export metadata ↓

Abstract

Long-horizon robot execution requires a clear distinction between a model's proposal, a controller's termination, and verified task completion. We present RegenHarness, an evidence-gated robot-agent harness connecting task planning to heterogeneous robot skills. Its execution architecture couples a model loop for context-conditioned proposals with an agent loop for dispatch, observation, verification, commitment, and bounded recovery. Four role-isolated contexts separate planning, supervision, verification, and recovery inputs. Versioned memory distinguishes observed facts from accepted task progress, while an identity- and version-bound commit gate controls updates to trusted task state. The runtime combines duplicate-dispatch control, resource leases, and recovery budgets under explicit backend contracts, and checks the original user goal before reporting completion. To our knowledge, we are the first to introduce an evidence-gated recursive self-improvement (RSI) protocol for embodied robotic agents. Across missions, execution records motivate candidate changes to context rules, task templates, routing, and recovery policies; fixed regression checks and release authorization govern their acceptance; versioned rollout and rollback preserve configuration traceability. This RSI protocol revises the harness configuration without online model-weight updates or permission to weaken the commit gate. A real quadruped deployment documents voice-triggered warehouse navigation, panoramic inspection, visual analysis, message delivery, return, and spoken reporting through linked audio, images, trajectories, and receipts. A separate circuit demonstrates why completion depends on execution history rather than endpoint proximity alone. Together, the cases demonstrate integrated perception, physical execution, communication, and history-dependent completion in real-world robot tasks.

Explore related subjects

cs.RO

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Kailin Wang, Haoxiang Jie, Yaoyuan Yan, Zhiyou Heng, Zhaosong Li. 2026-09-23. RegenHarness: A Robot Agent Harness with Evidence-Gated Recursive Self-Improvement. https://arxiv.org/abs/2609.27612

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

A Scalable Multi-Robot Framework for Decentralized and Asynchronous Perception-Action-Communication Loops

We develop a decentralized Perception-Action-Communication (PAC) system for multi-robot teams that enables them to collaborate in large scale, outdoor environments. Our system natively supports deployments at any scale by leveraging a graph neural network (GNN) to diffuse information hop-by-hop across the fleet's network. This achieves global collaboration from individual robots limited to local sensing and communication. Fully asynchronous, the core modules of PAC: perception, inter-robot communication, message aggregation and action are clocked at different frequencies with information flowing between them through buffers. We implement the PAC system as a series of highly extensible ROS2 nodes to serve as the foundational infrastructure for deployable swarm systems. PAC is validated in the real world with outdoor experiments with up to N=20 quadrotor robots and in simulations based on real-world data with up to N=100. These validations show that our system upholds crucial properties for field-deployable robot collectives: scalability, resiliency and repeatability.

cs.RO↗
arXiv2026

DexWrist: A Robotic Wrist for Constrained and Dynamic Manipulation

Development of dexterous manipulation hardware has primarily focused on hands and grippers. However, these end-effectors are often paired with bulky and highly stiff wrists that limit performance in human environments. More recent designs have adopted backdrivable actuation, but are still difficult to model and control due to coupled kinematics or high mechanical inertia from heavy links. We present DexWrist, a compact robotic wrist combining quasi-direct-drive actuation with a decoupled parallel kinematic mechanism to advance manipulation in highly constrained environments and enable dynamic, contact-rich tasks. It delivers 3.75$\pm$0.05 Nm rated torque, 0.33$\pm$0.06 Nm backdrive torque, 10.15$\pm$1.34 Hz torque bandwidth, $\pm 40^\circ$ ROM in both DOFs, and a diagonal velocity-constraint Jacobian (one-to-one motor-to-DOF mapping) in a 0.97 kg package. In practice, these properties increase workspace in clutter and stabilize contact without finely tuned admittance control. We evaluate DexWrist as a drop-in upgrade in simulation and on three robot arms across constrained and contact-rich tasks. In learned policy evaluations on the AgileX PiPER and UR3e, DexWrist achieved 50-76% relative improvements in success rate and reduced autonomous task completion times by 3-5x; on a torque-controlled Franka FR3, where a strong joint-impedance baseline already succeeds, it still completed the task 1.4x faster. Project page and videos: https://martinpeticco.com/dexwrist

cs.RO↗
arXiv2026

Imagine2Act: Leveraging Object-Action Motion Consistency from Imagined Goals for Robotic Manipulation

Relational object rearrangement (ROR) tasks (e.g., insert flower to vase) require a robot to manipulate objects with precise semantic and geometric reasoning. Existing approaches either rely on pre-collected demonstrations that struggle to capture complex geometric constraints or generate goal-state observations to capture semantic and geometric knowledge, but fail to explicitly couple object transformation with action prediction, resulting in errors due to generative noise. To address these limitations, we propose Imagine2Act, a 3D imitation-learning framework that incorporates semantic and geometric constraints of objects into policy learning to tackle high-precision manipulation tasks. We first generate imagined goal images conditioned on language instructions and reconstruct corresponding 3D point clouds to provide robust semantic and geometric priors. These imagined goal point clouds serve as additional inputs to the policy model, while an object-action consistency strategy with soft pose supervision explicitly aligns predicted end-effector motion with generated object transformation. This design enables Imagine2Act to reason about semantic and geometric relationships between objects and predict accurate actions across diverse tasks. Experiments in both simulation and the real world demonstrate that Imagine2Act outperforms previous state-of-the-art policies. More visualizations can be found at https://sites.google.com/view/imagine2act.

cs.RO↗