Abstract: Aiming at the sliding problem when the smart robot grabs the object, in order to provide a sliding signal to the manipulator, a highly sensitive magnetostrictive tactile sensor was designed using the inverse magnetostrictive effect of Galfenol. According to the inverse magnetostrictive effect, Euler-Bernoulli beam structural mechanics and Hook's theorem, the friction test model of the tactile sensor was established. The sensor was used for sliding detection, and the collected sliding signal was subjected to discrete wavelet transform to analyze the DWT detail coefficient of the sliding signal under different gripping forces.The threshold was set to 0.05 to control the robot to grasp the object.The results indicate that when the bias magnetic field is 4.2 kA/m, the applied friction is 4 N, the peak value of the sensor output voltage is 256 mV, and the sensitivity reaches 64 mV/N. The sensor is mounted on the robot and provides a stable control signal for the robot.
Key words: magnetostrictive; slide sensor; slide identification; discrete wavelet transform