Robotics · 2026-10-15

Towards Agentic Robots: Model, Data, Environment, and Harness

Chonghan Qin, Dongming Wu, Kaituo Feng, Jiaxu Wang, Xinlei Yu, Peiwen Sun, Yijia Fan, Xilin He, Bingao Chen, Weitong Lian, Ryan Ji, Wenqi Ren, Xiaobin Hu, Dahua Lin, Yu-Gang Jiang, Xiangyu Yue

arXiv:2610.20442 PDF Code

AI-assisted summary based on the paper. Consult the linked source for methods and limitations. Benchmark numbers appear after editor verification.

TL;DR

The authors survey and organize “agentic robots,” where agents autonomously plan and coordinate task execution, monitor progress, interpret execution feedback, and recover from failure. They trace evolution from robot policies to agentic systems and propose a taxonomy spanning four functional domains: Model, Data, Environment, and Harness. They also outline open challenges and directions for future research.

Why it matters

The paper argues that existing work on agentic robots is fragmented and lacks a consistent way to combine agentic functions into complete robot systems. By defining scope and providing a unified Model/Data/Environment/Harness taxonomy, it clarifies how agent decisions shape learning, data collection, world construction/evaluation, and runtime orchestration for capability improvement and reliability.

Method

Define agentic robots as systems where agents autonomously plan and coordinate task execution using observations and execution feedback, deciding steps and revisions without specifying every step.
Trace evolution from conventional robot policies to vision-language-action models, world models/world-action models, and finally agentic decision-making that reasons beyond learned policies.
Organize existing research into a taxonomy of seventeen categories across four domains: Model, Data, Environment, and Harness.

Limitation

The authors state that “current studies are not well connected” and that “researchers have no consistent view of how to combine different agentic functions into a complete robot system.”

Abstract (from arXiv)

Foundation models are driving a paradigm shift in robotic intelligence, moving beyond low-level action prediction to goal-directed orchestration of data collection, model training, task planning, and real-world execution. We characterize this emerging paradigm as an agentic robots, in which agents reason over human instructions, interact with robotic environments, rectify execution errors autonomously, and dynamically adjust operational strategies. Although this field develops fast, current studies are not well connected. Researchers have no consistent view of how to combine different agentic functions into a complete robot system. This paper provides a unified overview of existing research on agentic robots. We first trace the development from robot policies to agentic systems, covering vision-language-action models, world models, and world-action models. We then clarify the definition and scope of agentic robots and organize existing research into seventeen categories across four functional domains: Model, Data, Environment, and Harness. Across these domains, we examine how agents shape model learning and robot decisions, collect and curate robot data, construct worlds, tasks, and tests for learning and evaluation, and organize capabilities for robot use. Finally, we discuss open challenges and outline directions for future research. Looking forward, we hope this work contributes to a clearer understanding of agentic robots and supports new approaches to making them more capable, reliable, and adaptive.

Publication record

Bibliographic evidence describes where and when a record appeared. It does not validate scientific claims.

First public release: 2026-10-14 · Indexed by Pub-AI: 2026-10-15 UTC

preprint