Real-time obstacle avoidance for manipulators and mobile robots
International Journal of Robotics Research
A practical approach to motion-planning for manipulators with many degrees of freedom
The fifth international symposium on Robotics research
Robot motion planning: a distributed representation approach
International Journal of Robotics Research
A random sampling scheme for path planning
International Journal of Robotics Research
Robot Motion Planning
SGMIT: using selfish gene theory to construct mutualinformation trees for optimization
Proceedings of the first ACM/SIGEVO Summit on Genetic and Evolutionary Computation
Hi-index | 0.00 |
This paper addresses the path planning problem for manipulators. The problem of path planning in robotics can be defined as follows: To find a collision free trajectory from an initial configuration to a goal configuration. In this paper a collision-free path planner for manipulators, based on a local constraints method, is proposed. In this approach the task is described by a minimization problem under geometric constraints. The anti-collision constraints are mapped as linear constraints in the configuration space and they are not included in the function to minimize. Also, the task to achieve is defined as a combination of two displacements. The first displacement brings the robot towards to the goal configuration, while the second one allows the robot to avoid the local minima. This formulation solves many of classical problems found in local methods. However, when the robot acts in some heavy cluttered environments, a zig-zaging phenomenon could appear. To solve this situation, a graph based on the local environment of the robot is constructed. On this graph, an A* search is performed, in order to find a dead-lock free position that can be used as a sub-goal in the optimization process. This path-planner has been implemented within SMAR, a CAD-Robotics system developed at our laboratory. Tests in heavy cluttered environments were successfully performed.