Dynamics and Control of Robot Manipulators
Our research goal is to develop provably correct control algorithms for robot arms which enable dexterous tasks previously considered impractical or infeasible. Of particular interest are adaptive control algorithms for force control which will enable a robot to "learn" unknown parameters of both the arm itself and its environment to enable it to provide high-performance position and force tracking for manipulation tasks. Our methodology is to address these fundamental enabling theoretical issues for the next generation of robot arms, and to experimentally verify my research results in actual working systems.
Model-Based Adaptive Force Control for Robot Manipulators
We are working on adaptive control algorithms for force control which will enable a robot to "learn" unknown parameters of both the arm itself and its environment to enable it to provide high-performance position and force tracking for manipulation tasks. This research is supported by the NSF under CAREER Award BES-9625143. This work involves an active collaboration between our lab and industrial robot companies
High Performance Manipulator Design
High performance arm control is limited by the intrinsic mechanical capability of an arm design. DSCL Ph.D. student Jaydeep Roy has developed an extremely high performance semi-direct drive robot arm for control system experimentation. This arm, shown in Figure 5, provides torque-to-weight and structural vibration frequencies with approximately an order of magnitude improvements over present-day industrial arms. This research is supported under CAREER Award BES-9625143.
Design and Control of Medical Robotic Manipulators
Medical robotics has proven to be an interesting and challenging problem domain for both manipulator control system design and arm mechanical design. In collaboration with former DSCL post-doctoral student Dan Stoianovicci (presently Assistant Professor the Johns Hopkins Medical School), and collaborators Dr. Louis Kavoussi, Russel Taylor, Pat Jensen, and Dr. Eugene deJuan, we have developed several robots for surgical applications. Two devices have been patented by Johns Hopkins University. This research is supported by the NSF under grant IIS9801684, by the NSF Engineering Research Center for Medical Robotics and Computer-Assisted Systems and Technology EEC-9731748, and by direct support from medical instrument corporations.