编织型气动人工肌肉迟滞现象建模与应用

气动人工肌肉被广泛应用于柔性机器人、仿生机器人等多种机器人研究领域。但气动肌肉具有很强的非线性与明显的迟滞现象，直接限制了气动肌肉的精确控制。针对气动肌肉动力学建模问题，建立包含理想项与非线性迟滞项的完整气肌模型。通过仿真分析证明加入滞回模型后气肌的精度得到了提升：去程时精度提高了10%，回程时精度提高了33%。进一步，将该模型应用于机器人力控制关节上。通过曲面轨迹跟踪试验证明力控制精度从±15 N提升到了±5 N，同时将回程段控制精度提高了40%，并有效解决了力控制关节回程输出力不准确的问题。为进一步提高机器人柔性加工和装配领域的力控制精度提供了理论依据。

Hysteresis Modeling and Application of Mckibben Pneumatic Artificial Muscles

Pneumatic artificial muscles (PAMs) are widely used in flexible robots, bionic robots and other robot fields.But the strong nonlinearity and the obvious hysteresis directly affect the precision control of PMAs. Aiming at the problem of PMAs dynamic modeling, a complete model of PMAs, which contains ideal term and nonlinear hysteresis term, is built. The model was simulated and validated that the accuracy of PMAs can be significantly improved:the go-accuracy has been increased by 10%, the return-accuracy has been increased by 33%. Further, the presented model was applied to the active contact force flange of robot. The trajectory tracking experiment results prove that the accuracy of force control increase from ±15 N to ±5 N, and model can provide an improvement of 40% in the return-accuracy of the force flange. Meanwhile, the inaccurate problem of the force flange return output force has been resolved. The research results establish the theoretical foundation for further improve the force control accuracy in robot flexible processing and assembly field.