线驱软体机器人视觉伺服的研究

由于线驱软体机器人的良好弯曲特性及对复杂空间环境和障碍物很好的适应性，使得其非常适合在非结构化复杂环境中工作。然而软体机器人的复杂特性使得传统的基于关节位置反馈控制存在很大误差，限制了其应用。本项目拟通过视觉反馈实现线驱软体机器人的精确控制。为了实现这一目标，项目将开展软体机器人建模、控制器设计等方面的研究，实现一套快速可靠的视觉伺服算法。针对软体机器人高冗余度特性，结合控制任务与目标研究其优化控制方法。针对软体机器人应用场景，研究其在复杂受限环境下的视觉控制问题。项目拟在完全考虑软体机器人的非线性动力学特性和软体材料的弹性等特性的基础上，通过非线性理论分析控制方法的稳定性和鲁棒性，最后通过真实软体机器人实验验证其实用性及有效性。项目的顺利完成将推动软体机器人视觉伺服的技术发展，推广软体机器人在诸如医疗手术等需要高精度控制的领域中应用，对提高软体机器人的智能性具有科学意义。

The cable driven soft robots own good bending properties and good adaptability for complex environment and obstacles , which make it ideally suit for working in an unstructured complex environment. However, there exists some errors when traditional joint position feedback control is applied. This is mainly due to the complex nature of soft robots and also limit its application.The project is aim to control the cable driven soft robot precisely through vision feedback. To achieve this goal, the project will conduct research on soft robot modeling, controller design and other aspects and finally develop a fast and reliable visual servoing algorithm. To cope with the high redundancy features of soft robot, we will design the optimal control method by combining control tasks and objectives. To deal with the working environments for soft robots, this project will also study visual servoing in complex constrained environments. By fully considering the nonlinear dynamics of the soft robot and flexible properties of the soft material, this project will conduct stability and robustness analysis through nonlinear control theory. Finally the effectiveness and performance of the proposed methods will be verified through real experiments. The successful completion of the project will promote the development of visual servoing technology for soft robot and spread the application for soft robot in the field where high-precision control is necessary, such as medical surgery and so on. This work also has scientific significance to improve the intelligence of soft robots.