Sensorless robot collision detection based on optimized velocity deviation

It is one of the main directions of robot development to improve the security of robot and realize the safe human-robot collaboration. Collision detection is its basic function. In this paper, a robot collision detection algorithm based on robot velocity deviation is innovatively proposed. Due to the delay of the robot servo system, the actual velocity lags behind the ideal calculation velocity. The delay of the digital low-pass filter is used to deal with the ideal calculation velocity so that it can be well adjusted with the actual velocity. In this way, the velocity deviation of the robot can be reduced and the detection sensitivity can be improved so as to avoid the interference of robot acceleration and deceleration to collision detection. Compared to traditional algorithms, the algorithm proposed in this paper does not require additional equipment, nor does it need to detect acceleration, current, or torque. And complex robot dynamics operation are also unnecessary. The algorithm is simple and can be used directly on ordinary industrial robots. The experimental results show that the algorithm can detect collision effectively, sensitively and accurately, controlling the robot to stop moving immediately to avoid damage to human and robot itself.