Font Size: a A A

Structure Analyses And Optimizations Of N-type Resistive And Full Flexible Capacitive Three-dimensional Force Tactile Sensor

Posted on:2016-06-23Degree:MasterType:Thesis
Country:ChinaCandidate:X WeiFull Text:PDF
GTID:2308330470957704Subject:Control Science and Engineering
Abstract/Summary:PDF Full Text Request
Tactile perception is an important basis for highly intelligent robot, but the machine haptic technology currently used far less than the machine vision technology, the main reason is that most of the existing tactile sensors can not meet the demand of tactile perception application. So developping high-performance tactile sensors have become an important research direction in the field of robotics. Force sensing is the basic function of tactile sensor, in response to the complex operating environment of the robot, the force tactile sensor is gradually evolving from rigidity and one-dimensional to flexibility and muti-dimensional. For rigid and semi-flexible structure and limit single-dimensional force detection ability in normal or tangential diretion of the existing tactile sensor, this paper mainly design, analyzes and verifies the new structure of resistive and capacitive flexible three-dimensional force tactile sensor based on the preliminary work. This main works are as follows:Firstly, the N-type resistive flexible three-dimensional force tactile sensor proposed by our research group is simulated and calculated through multi-physics finite element method. The relationship between normal force, tangential force and multidimensional force and output-resistance matrix of the sensor model are researched. Results demonstrate that the sensor model can efficiently detect magnitude and direction of the inputed three-dimensional force.Secondly, decoupling research on N-type three-dimensional force tactile sensor is implemented the by BP (Back Propagation) neural network. The experiment proves that the output of network can accurately reflect the the inputed three-dimensional force, the result demonstrates that BP neural network algorithm can accurately decouple the output signal of N-type fexible three-dimensional force tactile sensor.Thirdly, for higher sensitivity and better stability of the capacitive sensors, we propose a full-flexible capacitive three-dimensional force tactile sensor model and describe the principle of the sensor in detail, then force analysis of sensor sensitive unit is made based on potential theory.Fourthly, the multi-physics simulation with electrical and mechanical field of the full-flexible capacitive tactile sensor sensitive unit is implemented using finite element technology. The experiment verifies the theoretical results. According to less change in capacitance when normal force applied on the sensor, the sensitive unit structure of sensor is improved. The results demonstrate that the new structure can efficiently increase the detection sensitivity of normal force.Finally, combined with international research status and development direction of the robot tactile, the paper recommends the future work for the existing problems in the research.
Keywords/Search Tags:Flexible Tactile Sensor, Three-dimensional Force, Multi-physicsSimulation, Finite Element Method, BP Neural Network
PDF Full Text Request
Related items