Analytical solution of a five-degree-of-freedom inverse kinematics problem for the handling mechanism of an agricultural robot

Abstract

Computing joint commands in real time by solving an inverse kinematic (IK) problem is an important manipulation task that is commonly encountered in agricultural operations such as fruit harvesting, disease detection and leaf sampling. To date, different numerical and analytical approaches have been proposed, and many of them are able to take constraints into account. In this study, a set of analytical algorithms is investigated to quickly solve the constrained five-degree-of-freedom inverse kinematic problem. The constraints due to actuator limitations and mechanical design of the mechanism are utilised in generating the solution sets for joint variables. Also a suboptimal solution for the inverse kinematic problem within this set is derived. The proposed methods are compared with a typical numerical solution based on non-linear constrained optimisation. It is shown that the proposed analytical and suboptimal semi-analytical solutions can be found in a much quicker fashion as compared to the non-linear constrained optimisation approach. The algorithms are validated by the handling mechanism in a ground robot platform specifically designed to operate in strawberry orchards.