Invasive weed optimization of swing-up control parameters for robot gymnast

Abstract

In this paper Invasive Weed Optimization (IWO) was used to investigate the optimum values of the control parameters for the swing up of the Robogymnast. The Robogymnast is classified as an under-actuated triple link pendulum. It was developed to study control problems associated with inverted pendulums. Like most underactuated robots the Robogymnast was built to exploit the natural dynamics of the mechanism. It 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. Two motors located on joint 2 (hips) and joint 3 (knees) control the movement of the mechanism. Joint 1 (hands/arms) is firmly attached to a freely rotating high bar mounted on ball bearings. IWO is used to optimize the swing up motion of the robot by determining the optimum values of parameters that control the input sinusoidal voltage of the two motors. The values obtained from IWO are then applied to both simulation and experiment. Results showed that the swing up of the Robogymnast from the stable downwards position to the inverted configuration was successfully achieved.

Analysing Various Control Technics for Manipulator Robotic System (Robogymnast)

“…LQR gives a reduced rising/settling duration in comparison to PID, while PID reduces overshoot and error, as well as being simpler to implement. The two types of controllers show differences in outcomes due to their differing strategies for feedback gain matrix calculation [17,23,24]. MATLAB Simulink was utilised to determine system response for the Robogymnast, to improve controller design and assess the function of the controllers within a closed loop system.…”

Enhanced the Control Strategy of a Triple Link Robotic System (Robogymnast)

“…Within this, this research considers upswing in the triple-link, nonlinear Robogymnast [3]. This system presents an opportunity to apply and evaluate different control systems, as it offers underactuation and high levels of mechanical complexity [4]. Controlling an underactuated system is highly challenging, due to the fact that most systems of this type are not full-state feedback linearisable about a given point of equilibrium, potentially not offering small-time local controllability (STLC) [5].…”