Reliability analysis and synthesis of robot manipulators

In this paper, a probabilistic approach to the manipulator accuracy is introduced. The concept of manipulator reliability is presented and the reliability index is proposed as a statistical measure of the accuracy of manipulators. A reliability analysis and synthesis procedure of the manipulator accuracy was developed. All kinematic parameters of a manipulator are assumed to be independent random variables following normal distribution. The manipulator reliability is defined as the probability of the end-effector position and/or orientation falling within a specified permissible region (known as reliable region) which depends on the intended application of the robot manipulator. Techniques are also presented to compute this reliability by an analytical approach and a simulation technique. Reliability synthesis of the manipulator accuracy was formulated as a nonlinear optimization problem by taking cost minimization as objective function, reliability indexes as constraints and the tolerances of the kinematic parameters as design variables. The presented procedure helps designers to compute optimal assignments of manufacturing tolerances to individual kinematic parameters which give the desired manipulator reliability while minimizing the overall cost. Finally, a six-link manipulator is considered for numerical illustration and results from both analytical approach and simulation are compared.<<ETX>>