Abstract With the increasing integration of robots into various fields such as industry, agriculture, specialized industries, and home services, research on human-robot coexistence and collaboration has become more urgent, which is crucial for enhancing the operational capabilities of robots and ensuring personal safety. As a kind of soft actuator with smart structures, pneumatic artificial muscles (PAMs), known for their high power-to-weight ratio, ease of installation, clean energy usage, and resistance to electromagnetic interference, can form a series of PAM-actuated robots combined with rigid robotic architecture, endowing them with the ability to coordinate rigidity and flexibility. Recent advancements in the modeling and control methods of PAM-actuated robots have significantly improved their practical application value. However, there is a lack of systematic and structured review papers that explore the differences and connections among various modeling and control methods for PAM-actuated robots, and analyze the progression of their development. This paper will provide a detailed discussion of the key characteristics of various modeling and controller design for PAM-actuated robots, and offer guidance for the next generation of research directions in this field.
Zhang et al. (2025) studied this question.