In this paper the problem of swing-up control of an under-actuated, three-link robot gymnast (Robogymnast) is discussed. Robogymnast mimics the human acrobat who hangs from a high bar and tries to swing-up to an upside-down position with his/her hands still on the bar. Unlike human acrobats, Robogymnast's hands are firmly attached to a freely rotating high bar mounted on ball bearings. Although this helps during the swing-up phase, it poses a great challenge to balancing the robot at the upright position. The motion of Robogymnast was initiated by applying regularly changing sinusoidal torques to the two motors located respectively at its shoulder and hip joints. As the swing angle increases, the amplitudes of the applied sinusoidal torques were increased while their respective frequency was reduced proportionately. Experimental results showed successful swing-up of Robogymnast from the stable downwards position to the inverted configuration.
This paper 1 focuses on using the Bees Algorithm in both its basic and enhanced forms to tune the parameters of a fuzzy logic controller developed to stabilise and balance an under-actuated two-link acrobatic robot (ACROBOT) in the upright position. A linear quadratic regulator (LQR) was first developed to obtain the scaling gains needed to design the fuzzy logic controller. Simulation results confirmed that using the Bees Algorithm to optimise the membership functions and the scaling gains of the fuzzy system improved the controller performance.
The design of robust computer control systems for balancing and attitude control of double and triple inverted pendulums is considered in this paper. For the double inverted pendulum, a DC motor mounted at the upper hinge is used to balance and control attitude of the upper link. For the triple inverted pendulum a DC motor mounted at the middle hinge is used to control the middle link, whereas proportional position control applied to a motor at the upper hinge is utilised to maintain the upper link in alignment with the middle link. In both cases the lower hinge is left free to rotate. The controller designs are based on linearised discrete-time models of the inverted pendulums. Each controller utilises state feedback implemented via reduced-order state observers. The relative stability properties of the control systems are evaluated using Nyquist plots of suitably defined functions. The controllers are designed using Matlab and implemented in a PC using C language. Experimental results showed satisfactory performance.
This paper describes the control system for an eight-degree-of-freedom biped robot built at the University of Salford. The controller enables the robot to walk in the sagittal plane on smooth level terrain and is essentially an observer-based controller utilising state feedback, integral action and feedforward control. The robot posture is controlled by selecting constant reference set points for the control system. Locomotion is achieved by suitably modifying the reference set points. The robot walks with a step length of approximately 0.3 m and a speed of about 0.03 m/s.
Until now, many design and manufacturing tasks have only seen partial computerization. One such task is that of assembly planning. This paper presents a system for use in a subarea of assembly planning, the configuration of grippers for robotic assembly. The system is fully automated and thus represents a departure from current practice. The paper gives an example of the application of the system to simple gripper design problems.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.