Search

This research considers the design of a sensor system for walking robot which can be used for coverage path planning and precise surface exploration. The paper presents an updated mechanical design of a minimalistic walking robot with a fixed step size. The robot has only two degrees of freedom. The specific mechanical design prevents the accumulation of any odometry error and does not require any external navigation system. The robot is equipped with a variety of distance and tactile sensors which allow precise exploration of the surface and any high or low spots of the surface to be located. This paper proposes an approach for path planning and surface exploration by using the proposed sensor system. Despite that the robot is based on a minimalistic approach, it can move and operate on complex surfaces. Also, by altering its movement direction it can effectively use its sensor system for distant global object scanning or for local exploration of the walking surface. The conducted experiment showed that the presented in this research mechanical design of a walking robot with fixed step can be used for precise surface exploration.

This research is focused on trajectory planning for redundant robotic manipulators with constrained joint space. The redundant robots have infinite number of solutions of the inverse kinematics problem. Due to the constrained joint space, the number of sets of joint angles attaining given end-effector position are limited. This requires passing through singularities when trajectory movement is executed. Each pass through a singularity increases the execution time. This paper takes into consideration the characteristics of both the desired trajectory and the robotic manipulator. It proposes an algorithm which discretizes the desired trajectory and constructs a weighted directed graph which represents all the possible ways for planning the trajectory movement. The trajectory movement with the minimum execution time can then be planned by finding the path with minimum cost of the constructed graph. The applicability of the algorithm is evaluated through experiments.

Recently the interest for redundant robots is increasing because of their wider functional capabilities. Usually the main advantage of redundant robots is their capability to avoid obstacles. Another advantage of this type of robots is the ability to reach a point from their workspace via different configurations of the links which allows redistribution of the reaction forces along the kinematic chain. In this way the drive power can be optimized. In this work is considered a planar redundant 3D printed robot with three degrees of freedom. Its kinematic and dynamic descriptions are presented. The dynamic model is developed using the energy conservation law and graph theory. Finally experiments of its real motions, based on the dynamic model are presented graphically.