Human Support Robotics Based on Haptic Recognition

触觉模态是人机交互中最重要的考虑因素之一。因此，本课题开发了基于触觉传感器信号的命令识别技术。首先，开发了一种称为触觉手势的命令识别技术，用于人类支持机器人的有效操作。针对触觉传感器刚度和强度低的缺点，设计了一种多支撑点机构。实验结果表明，多支撑点机制提高了识别率。其次，针对非凸形末端执行器，提出了一种全身力觉算法。该算法克服了触觉传感器形状的限制。第三，提出了一种基于手指跟踪力信息的身份识别技术。与其他生物特征识别方法相比，由于力信息的不可见性，该方法具有安全性高的优点。

Haptic modality is one of the most important considerations in man-machine communication. Therefore, this project developed a technique to recognize commands based on the signals from tactile sensors.Firstly, the command recognition technique so-calledhaptic gesturewas developed for efficient operation of a human support robot. Since the tactile sensor has drawbacks of low stiffness and low strength, a mechanism with multiple supporting points was developed. It was shown that the recognition rate improves with the multiple supporting point mechanism.Secondly, a whole-body force sensation algorithm for a non-convex shaped end-effector was developed. The limitation on the shape of the tactile sensor was eliminated with this algorithm.Thirdly, this project introduced a personal identification technique based on force information of finger tracing. Compared to other biometric methods, the proposed method has the advantage of high security due to the non-visible nature of force information.