Robot arm manipulation using depth-sensing camera and inverse kinematics

Robotic arms can be controlled by human operators using different types of controllers or manipulators. For example, a Titan IV robot arm can be mounted on a ROUV (remotely operated underwater vehicle) for seafloor operation and can then be remotely controlled by a sophisticated manipulator, a “Master Controller”. An operator needs a lot of training with this type of controller before they could apply their skills manipulating real robot arms in the field. There are, however, simulators like GRI Simulations Inc.’s manipulator trainer which help a user train virtually on a particular robotic arm using either a master controller or a joystick. Compared to joysticks, master controllers are much more sophisticated and expensive devices. On the other hand, joysticks are not as convenient as controller mechanisms, since they are more generic products and do not map the functionalities of master controllers as well as the custom master controllers. This thesis presents a new technique to manipulate a robotic arm which uses an inexpensive depth-camera to capture the user input and inverse kinematics to define the motion of the robotic arm. Along with the easier manipulation of the robotic arm, the presented technique also adds some gesture commands to control the end-effector which makes the interaction more intuitive. To test the efficacy and efficiency of the proposed method, a user study was conducted in which 18 participants were asked to perform two placement tasks using a keyboard, a joystick and the depth-camera based interface. The presented technique is inexpensive and the results of the study suggest the technique is a good option for controlling robot arms with configurations similar to that of the Titan IV