Signal Processing of Robotic Sensitive Skin Based on DSP+FPGA

摘要

Sensitive skin is a new perceptual system for robot. It is mainly used for real-time collision avoidance for robot in unknown environment. The sensor signal will be interfered by the around noises, and it is difficult to meet the requirement of accuracy, real time and stability of the system when using traditional signal processing method. In this paper a novel digital signal processing system based on DSP+FPGA is discussed. The Fast Fourier Transform (FFT) and periodogram is used for analyzing the signal, and some methods are taken to rectify errors from the limitation of the FFT. The simulation and experiment validate that this method can reduce the effect of noises for system, and can satisfy the real-time collision avoidance for robot.