Research‎ > ‎

Path Planning for Robotic Manipulators

Control strategies used for robot manipulators can be classified in two types including “joint coordinates” and “Cartesian coordinates”. While in the ‘‘joint coordinates’’ approach, the desired joint positions are specified, in the ‘‘Cartesian coordinate’’ approach the desired end effector position or path is specified in Cartesian space. The ‘‘Cartesian coordinate’’ approach are considered as one of the challenging problems encountered in robotic field. In this approach, robot manipulators are needed to track a predefined path in Cartesian space or move to desired position in such a way that no collision with obstacles in the environment occurs. Therefore, feasible paths for each joint of robot must be determined in such a way that required criteria are met.

 

In this research, different methods of path planning for robotic manipulators are analyzed and compared. Using these algorithms, a feasible path for joints of manipulator are determined in such a way that end-effector of robot tracks predefined path or moves to goal point in Cartesian space while at the same time avoids collision with obstacles and singular configurations. 


Studied algorithms are as follow:

1) Roadmap, 2) Cell Decomposition, 3) Randomized Road Map, 4) Rapidly Exploring Random Tree, 5) Artificial Potential Fields, 6) Optimization Based Algorithms, 7) Task-priority Redundancy Resolution, 8) Gradient Projection, 9) Generalized Inverse Jacobin, 10) Extended Jacobin approache.