Zeroing neural network with fuzzy parameter for cooperative manner of multiple redundant manipulators

In this work, a distributed scheme based on quadratic programming for cooperative manner of multiple redundant manipulators with limited space constraints is studied. For the solution of the distribute scheme, a fuzzy parameter zeroing neural network (FPZNN) is presented for the coordinated motion planning of multiple redundant manipulators via absorbing fuzzy logic strategy. The fuzzy logic strategy is generated by fuzzy logic system (FLS), which can adjust the systematic parameter according to the convergent error. The scaling parameters included in the design of FPZNN are appropriately tuned with well-defined FLS output values. Theoretical analyses prove that in the absence of noise, all the manipulators can reach their desired position after the given tasks no matter for the initial joints position with an exponential position convergence. Furthermore, FPZNN with sig-bi-power activation function (SBP-AF) is applied to solve the distributed scheme of multiple redundant manipulators and obtain finite convergent rate for position error of each joint. In addition, comparative experiments prove that the FPZNN model has superior performance for the solution of the distributed manner with multiple redundant manipulators compared with the corresponding ZNN model combining with different AFs.