TDPN based Control for Haptic Setup

In the proposed system, a robotic avatar is being manipulated by a human operator from a remote distance. This requires the velocity and force values to be sent to the robot, which is how the robot has to move. In the system, an energetic connection is established between the robot and human operator. Through software, the power and energy leaving and entering the communication channel are calculated. Ideally, the powers at the human operator port and the robot port are the same at any point in time, this we call full transparency. However, the channel introduces some time delay in the system. This may introduce energy generative, or non-passive, elements to the system, which destabilizes the whole system.

A controller has to be designed in order to dissipate this generated energy. A solution can be to apply constant damping to the system. This method was proved to guarantee passivity, but has a large energy loss. A better system would be one which dynamically determines when energy is being generated and applies an appropriate damper based on the amount of energy generated.

This thesis proposes using the time-delay-passivity-control (TDPN) method to solve this issue. This method looks at the energy flowing into one port of the system and out of the
other port of the system. The system has to calculate the energy going into the channel, send that information over said channel, and calculate the outgoing energy. It then has to calculate the difference between the current outgoing energy and the received energy which went into the system. The latter is received with some time delay introduced by the channel. If the difference is greater than zero, the system must be passive. If not, a damper has to be put into the system. This damping constant has to be calculated on the basis of the current value of the energy difference. Once the system is passive again, the constant should become zero once more.