Position and orientation sensor using multiple video cameras

The positioning of robotic manipulators in world coordinates is performed as an open loop control function, as measuring the gripper position in world coordinates is difficult. A 4×4 matrix [P] can be used to describe the position, orientation and approach vectors of the gripper of a manipulator. A 4×4 transformation matrix [T] can be produced which will enable the [P] matrix to be determined for any manipulator configuration. [P]=[T][B] where [B] is the 4×4 matrix representing the base position. This constitutes the forward control path. A world coordinate position sensor has been developed using the cameras and can be used to provide the feedback path and close the positioning control loop. The actual position of the gripper which is measured by the sensor and is represented by [P¹] may be different from the predicted position by the error [E] error, [E]=[P]-[P ¹]. The sensor enables the error [E] to be reduced to zero. The position and orientation sensor is implemented using a single FPGA. It performs (almost) to the required specifications, enabling the position of the point source of light to be determined, and hence the position and orientation of the end effector to be calculated