Integrated vision and force control of a 3-DOF planar robot

Due to task requirements, robotic arms often need force control to accomplish different tasks such as assembling, grasping, and wiping that demand compliant motions. Thus, integrating vision with force sensor to achieve hybrid force/position control is a challenging research topic. This paper is devoted to the problem of precisely controlling the pose of a three degree-of-freedom planar robot for a class of constrained motions. The goal of this research is to integrate vision with force sensor in a simple way that mimics human behavior. The main idea is to maintain the contact force while controlling the pose of the end-effector with vision. A planar robot is considered to apply a fixed contact force against the environment, straight/curved surface, using a single camera mounted above. The proposed system utilizes a force sensor to measure the contact force and employs Cartesian-based encoding to encode the visual task. Two approaches, subspace and hierarchical, are employed in the research. Both can guarantee to reach the control goal with precision. In the subspace approach, the control design is based on the decoupling the vector space of control variables into two orthogonal subspaces using observed visual data. As for the hierarchical approach, a distinct and simple way is proposed by controlling the pose of the end-effector and the contact force in a hierarchical loop structure. These control strategies are verified to be effective through simulations and experiments with an industrial robot.