Neural Network Control of Robot Arms

This is a good project report on Neural Network Control of Robot Arms. This is good seminar topic for Mechanical and even electronics students. It tells about the controlling of a robotic arm along a given trajectory. Controlling a robot along a given trajectory can be divided into two sub-tasks. The first one is an inverse kinematic calculation which transforms the trajectory, usually specified in a Cartesian coordinate, into a sequence of required joint positions. The second one is the generation of joint torques from required joint angles and their derivatives. For intelligent robots the inverse kinematics calculations have to be calculated in real-time which is a time consuming computational procedure. You can also Subscribe to FINAL YEAR PROJECT'S by Email for more such Projects and seminar topics.

Recommended Project: Robotic Manipulator Arm Mechatronics

Our aim is to recover the inverse kinematic relationship of a robot using a learning algorithm with a set of exemplars consisting of joint angles and their corresponding end-effector coordinates. Once the relationship is recovered and implicitly stored in the NN, it can be used to calculate joint angles corresponding to a given world coordinate much faster than by the raditional way.

DIRECT DOWNLOAD

Advertisements:-