Research On Grasping Strategy Of Manipulator Based On Active Excitation Tactile Sensor

Soft grasping is a hot spot in the field of robotics research.It is a comprehensive discipline that integrates sensors,control,and artificial intelligence.There are many research results on soft grasping strategies at home and abroad,but the judgment of whether the grasping is stable is mostly based on sliding state detection.However,the sliding state signal is unstable and has strong randomness,which is easy to cause misjudgment,and the signal is complicated.The processing and analysis requires a large number of complex calculation processes,resulting in poor real-time performance and poor soft-grabbing effects.Aiming at these problems in the current soft grasping research,this paper creatively proposes a soft grasping strategy based on active stimulation of the tactile sensor system.With the purpose of capturing the stable and continuous signal,and based on the principle of acoustic resonance spectrum,an active excitation tactile sensing unit was developed.The sensing unit can actively generate tactile signals and the signal is continuous,which is essentially different from traditional tactile sensors.Taking real-time signal acquisition and control as the research purpose,this paper designs an active excitation tactile sensor system with Compact RIO as the control core,and uses Lab VIEW as the software development environment to design three-level programs of FPGA,RT and PC.With the soft capture strategy as the research purpose,time-domain standard deviation analysis,frequency-domain FFT analysis,time-frequency domain wavelet packet analysis,EMD analysis and HHT analysis were performed on the collected signals of the capture process.The law that stimulates the grasping process of tactile sensing is that the system is relatively stable in the two states of grasping and grasping,and the system is the most unstable during grasping.Based on this,three grasping strategies are proposed: soft grasping strategy based on frequency band energy analysis;Soft grab strategy based on FFT spectrum analysis;Soft grab strategy based on time-domain standard deviation analysis.These grab strategies are fundamentally different from traditional grab strategies.The traditional grab strategies based on tactile sensor signals mainly focus on the sliding state Signal,according to the sliding signal to determine whether to add force,and the soft grasping strategy based on the self-made active stimulation haptic signal in this paper is to judge whether the force needs to be added and whether the system is stable according to the change trend of two stable states(grabbing and grasping).The experimental results show that the grasping success rate of apples,500 g weights and foam is more than 90%,and the grasping success rate of raw and cooked eggs can also reach more than 50%.The experiment verifies the feasibility of the soft grasping strategy.This paper also does a system adaptability experiment,which proves that the system can be used in real life production after further improvement.