Scene perception and motion planning through robotic vision

The use of robotic vision is extensive in any field. This thesis aims to apply the theoretical concepts of the coverage strength model in order to achieve a task using robotic vision. The task mentioned is motion planning for the robotic arm such that it does not collide with an obstacle while performing its operation. This research can be applied to robots in an industrial work cell or an unmanned system. Hand in eye configuration is used to control the robotic motion. Different levels of calibration are implemented, as a camera network is involved. An algorithm is used to implement the motion planning. This algorithm also involves the pose estimation of the objects used in the work cell. The work cell is modeled with a camera network and a robot. The calibration helps to visualize the position of the various entities of the work cell. The algorithm is applied to the physical experimental setup and results are recorded