Abstract: A novel three limb robot was described and two motion planning methods were discussed. After the introduction of the robot mechanical structure and the human-robot interface, method A is on the basis of the genetic algorithm. A kind of variable structure genetic algorithm is proposed in method A to solve the problem of motion planning of the three limb in dynamic environments. The variable structure genetic algorithm changes the original structure by abandoning Elitist Model, expectation selection, reproducing population and changing the probability of crossover and mutation. Experiments results show that the former algorithm is effective in static environments and the latter algorithm is good at dynamic environments. Method B is a two-grade search mechanism. The first-grade search method using genetic algorithm tries to find an optimized target position and orientation of the three limb robot. The second-grade search method using virtual compliance tries to avoid the collision between the three limb robot and obstacles in a dynamic environment. Experiment shows the feasibility of the two-grade search mechanism and proves that the proposed motion planning method can be used to solve the motion planning problem of the redundant three limb robot without deficiencies of traditional genetic algorithm.