scene organization using robotic vision - phython developer

Robots play a major role in every part of life. They might be assistants at home or an appliance or work in an industrial environment. Dynamic obstacles can be found in any kind of work cell of a robot. Collision avoidance is a very important aspect in an industrial robotic cell. In a dynamic environment obstacle detection plays a very important role. In this thesis work the concepts of visual feedback is used .In a normal pick and place application of an industrial robot there might be dynamic obstacles along with the object to be picked. The approach taken for this thesis can be compared to a pick and place application where the target objects are in continuous movement in a conveyer belt. The robot’s path has to be changed in order to avoid the dynamic obstacles. Dynamic or static obstacles can disrupt the operation of any robot. The robot has to find a way to deal with the occlusions of the target object caused due to dynamic obstacles. It is necessary for the robot to perceive an obstacle in order to prevent collision. This is done using a camera network. The robot is in continuous movement. The robot must reach the target to grasp it or for any other application when there are no dynamic obstacles which would cause collision between the obstacle and robotic arm. Once occlusion is detected the robot must either stop moving or change its path.