Human-Robot Interactive Control
This research proposes to develop a new control strategy, a software approach, in the area of human-robot cooperation: A human-robot interactive control that enables robots and humans to work together in the same workspace. The important issue of human-robot interaction is how to control and plan a robotic operation such that the human and the robot can cooperate in a complementary and safe manner. For example, in certain tasks, when a human temporarily holds the robot’s arm, the system in motion should remain stable and should not accumulate torque. The key concept of the proposed controller is based on acceleration feedback, passive interactive control and time independent path planning with actual position feedback. Without accelerometer applied, the joints’ accelerations are obtained via twice differentiated of joints angle output from optical encoders. Without force sensor and motion detector applied, the computer determines the robot and human interacting via software computations from the reading change in the joint’s encoders. By using active torque control with joints acceleration feedback, the inertia matrix of the system can be compensated and the whole system becomes lighter than it really is. For an unknown system, the unknown nonlinear terms can be computed from the acceleration measurement, and it is used as feedforward term to linearize the system dynamic. Passive interactive control ensures that humans can safely hold the manipulator’s arm. With stability and time independent path planning, it guarantees that error will not be increased when a robot is held. Furthermore, the proposed controllers provide a new idea for obstacle avoidance and for contact with unknown environments. This dissertation includes the development of the proposed control strategy, simulation studies, and experimental validation of the concepts using a 4 Degree-of-Freedom Adept 604A SCARA manipulator.
School Location:USA - Ohio
Source Type:Master's Thesis
Keywords:human robot interaction inertia compensation acceleration feedback time independent path planning obstacle avoidance
Date of Publication:01/01/2003