Tham khảo tài liệu 'mobile robots - moving intelligence part 3', kỹ thuật - công nghệ, cơ khí - chế tạo máy phục vụ nhu cầu học tập, nghiên cứu và làm việc hiệu quả | Combined Torque and Velocity Control of a Redundant Robot System 71 the smaller circles show obstacles detected with ultrasonic sensors. F denotes the force of the obstacle that collides with the platform. The collision place is only approximate because the obstacle position and force were not measured. The manipulator moved away from the obstacle after the collision in s. The impact force was approximately 4 N. Fig. 21. Configuration of the mobile manipulator before and after the collision. Motion in contact with the environment One part of the trajectory reference large circle was inside of the wall. The mobile manipulator end-effector was tracking the trajectory to the wall and than it was pressing to the wall. The force of the contact is not controlled directly. It is dependent on the tracking error and the manipulator configuration. The motion of the manipulator in contact is shown in Fig. 22. Fig. 23 shows the force on the wall. Force measurement was used only for validation and not for the control. Fig. 22. Motion of the mobile manipulator in space with obstacles in contact with the wall. 72 Mobile Robots moving intelligence Fig. 23. Normal component of the manipulator force on the wall. 6. Conclusion The use of the robots in the real world mostly depends on the development of a autonomous robotic systems that integrate mobility and manipulability. This chapter has presented a new approach to the robot control. The control is called combined control since it combines integrates two different types of robot control . the torque and the velocity control in a single robot system. Later on the mathematical analysis of the proposed combined control has been shown. Verification has been done on a simulated and a real system composed of a velocity controlled mobile platform and a torque controlled manipulator. The results show good tracking of the desired trajectory in the task space and in the null space depending on the controller gains. Controller .