Task scheduling with obstacle avoidance for industrial manipulators operating in 3D environments
In this paper a new method is introduced to solve the scheduling problem for a point-to-point manipulator motion in three-dimensional (3D) environments. The robot’s 3D environment is cluttered with obstacles of arbitrary size, shape and location, while a set of task-points are located in the robot’s 3D free-space. The objective is to determine the optimum collision-free robot’s tip tour through all task-points passing from each one exactly once and return to the initial task point. This scheduling problem combines two computationally hard problems: the task scheduling problem and the collision-free motion planning between the task points.
In the proposed approach, the Bump-HyperSurface concept is used to represent the entire robot’s 3D environment through a single mathematical entity, a 3D B-Spline surface. Then a Genetic Algorithm (GA) with a special encoding is introduced to take into account the multiple solutions of the Inverse Kinematics. The results show that the method can determine the optimum sequence of a considerable number of task points in complex 3D environment for a PUMA manipulator.
Dear Sir,
In this paper, a PUMA 560 has been used for the experiments. This is a robot of six degrees-of-freedom and all rotational joints. For this robot, as with many industrial robots, the joint axes of joints 4, 5 and 6 all intersect at a common point and this point of intersection coincides with the origin of frames {4}, {5} and {6}. Thus, these three joints cannot be represented, since they all lie in the wrist of the manipulator. Despite the fact that four links are represented, the robot is encountered as a 6-dof manipulator having 8 possible configurations at each point.
Regards
Dear Author,
Thanks for your clarification. I understand from your explanation that the model used is the simulation was constructed using 4-rotational link robot with 6-DOF which can have 8-different configurations. Is this correct?. If so, please explain how the 8-different configurations are represented in this case.
Regards
Dear Sir,
As you have understood, the link representation does not constrict the degrees of freedom. This is only an issue of depiction. Concerning the multiplicity of configurations, the PUMA can reach certain goals with 8 different configurations. These configurations are defined by solving the Inverse Kinematics Problem. More specific, for a certain pose (position and orientation) of the end-effector, 8 sets of the joint angles (θ1, θ2,.. ,θ6) can be defined. The algorithm chooses the appropriate configuration corresponding to each task-point and the robot representation is based on the corresponding set of joint angles.
I hope I have answered your query. If not, please do not hesitate to ask me again.
Regards
Dear Author,
Thank you for your clarification. I can see now that the drawing in the paper is used only for depiction of the robot tracking the path points, while not showning all of the possible configurations details, specially that related to the robot wrist. Thanks for your explaination.
Regards
Dear Authors,
Thanks for your paper. Would you mind explaining the DOF involved in the simulation?. I can see only 4-Link robot while most of the discussion were about full 6-DOF PUMA robot. Also, the 8-states of configurations for the PUMA robot are related to having different orinetations related to the 6DOF structure of the PUMA. Would you mind explaining the related configurations in your simulations.
Regards