A Modular Software Architecture for Safe and Scalable Mobile Manipulation Systems

Robotic manipulation systems are increasingly deployed in real-world environments where reliability, safety, and scalability are as critical as task performance. As these systems grow in complexity, software architecture has emerged as a primary determinant of operational robustness and long-term maintainability. This paper presents a modular software architecture for mobile manipulation robots that emphasizes separation of concerns, explicit task lifecycle management, and event-driven coordination under real-time constraints. The proposed architecture decomposes robotic functionality into layered subsystems spanning perception, task reasoning, motion and skill generation, and execution and control. Design choices are motivated by the need to manage heterogeneous time scales, partial failures, and safety-critical behaviors. The architecture is evaluated through multiple case studies, including a holonomic mobile base with a specialized cleaning end effector, warehouse automation systems, and assistive robotics platforms. The results demonstrate that disciplined architectural design improves fault containment, system observability, and deployment reliability, supporting scalable robotics development and safe operation in dynamic environments.