Agentic AI for Robot Control: Flexible but still Fragile

TL;DR

An agentic control system in which a reasoning-capable language model plans and executes tasks by selecting and invoking robot skills within an iterative planner and executor loop is presented.

Abstract

Recent work leverages the capabilities and commonsense priors of generative models for robot control. In this paper, we present an agentic control system in which a reasoning-capable language model plans and executes tasks by selecting and invoking robot skills within an iterative planner and executor loop. We deploy the system on two physical robot platforms in two settings: (i) tabletop grasping, placement, and box insertion in indoor mobile manipulation (Mobipick) and (ii) autonomous agricultural navigation and sensing (Valdemar). Both settings involve uncertainty, partial observability, sensor noise, and ambiguous natural-language commands. The system exposes structured introspection of its planning and decision process, reacts to exogenous events via explicit event checks, and supports operator interventions that modify or redirect ongoing execution. Across both platforms, our proof-of-concept experiments reveal substantial fragility, including non-deterministic suboptimal behavior, instruction-following errors, and high sensitivity to prompt specification. At the same time, the architecture is flexible: transfer to a different robot and task domain largely required updating the system prompt (domain model, affordances, and action catalogue) and re-binding the same tool interface to the platform-specific skill API.

Figures (2)

• Figure 1: Robotic platforms used in our experiments (see Section \ref{['sec:experimental_evaluation']}): the Mobipick mobile manipulator (top), operating in the object collection scenario, and the Valdemar robot (bottom), operating in an agricultural environment.
• Figure 2: Agent interaction graph (router $\rightarrow$ chatbot or OpenAI o3 based planner/executor $+$ Goal completion critic).