An integrated process for design and control of lunar robotics using AI and simulation

2025-09-15

AI Summary

Co-optimizing mechanical design and control strategies for lunar construction equipment remains challenging due to the lack of integrated development frameworks. Method: This paper proposes an integrated co-design workflow leveraging high-fidelity, real-time 3D simulation to concurrently optimize mechanical structure and autonomous control policies. It introduces OpenPLX—a readable/writable declarative language—to unify CAD modeling, multibody dynamics, robot–regolith interaction, and non-ideal sensor simulation. The framework integrates a physics engine, vision-language navigation models, and reinforcement learning–based motion controllers. Contribution/Results: Evaluated on two representative lunar robotics cases, the workflow enabled the development of a lunar rover capable of natural-language instruction understanding and autonomous obstacle negotiation. Experimental results demonstrate significant improvements in mission execution efficiency under complex lunar terrain conditions and accelerated R&D iteration cycles.

Technology Category

• Intelligent Robots: Learning & Optimization for ROB
• Planning, Routing, and Scheduling: Planning with Language Models
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work

Abstract

We envision an integrated process for developing lunar construction equipment, where physical design and control are explored in parallel. In this paper, we describe a technical framework that supports this process. It relies on OpenPLX, a readable/writable declarative language that links CAD-models and autonomous systems to high-fidelity, real-time 3D simulations of contacting multibody dynamics, machine regolith interaction forces, and non-ideal sensors. To demonstrate its capabilities, we present two case studies, including an autonomous lunar rover that combines a vision-language model for navigation with a reinforcement learning-based control policy for locomotion.

Problem

• Integrating AI and simulation for lunar robotics design
• Developing autonomous control for lunar construction equipment
• Linking CAD models to high-fidelity lunar environment simulations

Innovation

• AI and simulation integrated design process
• OpenPLX declarative language linking CAD
• Reinforcement learning control for locomotion