We present a comparison of a LQR controller and Q-learning on a simulation of a triple inverted pendulum. While the LQR controller was able to balance the pendulum, Qlearning was unable due to memory and computing limitations. We examine the strengths and weaknesses of each method and compare their abilities.

This paper investigates the design and analysis of three controllers used to stabilize an inverted pendulum on a cart. This is accomplished by decomposing each control algorithm into two separate phases: swing-up control and stabilization control. A classical, H2, and H∞ design are each considered to stabilize a nominal linearized model of the inverted pendulum on a cart in the upward vertical ...

Model Predictive Control (MPC) is used to improve the performance of energy control for swinging up a pendulum. A new MPC method is developed in continuous time, but it explicitly considers its digital implementation letting the control signal be piecewise constant. The stability properties of the algorithm are analyzed in terms of the free MPC design parameters. The achieved performance improv...

The inverted-pendulum system is a favorite example system and lecture demonstration of students and educators in physics, dynamics, and control. This system is a simple and valuable laboratory representation of an unstable mechanical system. This paper traces the early history of the inverted-pendulum system, and also compares several of the early treatments from the literature between 1960 and...

We design a nonlinear control law for a four degree of freedom spherical inverted pendulum based on the forwarding technique. We first explore the forwarding structure of the spherical inverted pendulum model and then find a control law to stabilize the angle variables. Next, we develop a nested saturating controller for the whole system. The control law is evaluated through simulations.Copyrig...

In this article is the automatized procedure for deriving of n-link inverted pendulum motion equations presented. Example of 2 link inverted pendulum is included. The LQR algorithm using Maple input equations is proposed. Also the comparison between SimMechanics and Simulink is included.

In this paper, a novel search operation is proposed for the neuroevolution of augmented topologies, namely difference-based mutation. This operator uses differences between individuals in population to perform more efficient optimal weights and structure model. The difference determined according innovation numbers assigned each node connection, allowing tracking changes. implemented algorithm ...

We revisit a previous high-low gain control idea for a 4 DOF spherical inverted pendulum using a different approach, inspired by a nested saturation tool proposed by Marconi and Isidori, that provides explicit tuning rules to deal with certain bounded external disturbances. The update controller is a robust, decentralized and “global” controller.

Balance control of a simulated inverted pendulum attached to a circular base is presented. This type of platform is analogous to a biped robot with circular soled feet in single support phase. Circular feet have been shown to be more energy efficient than flat feet during walking, and in this paper we present another advantage, where circular feet do not suffer from ground separation when apply...

Exact feedback linearization converts any system into appropriate form such that the controller design becomes convenient for the system. In this technique, control algorithm is developed based on exact feedback linearization through energy control. This approach is applicable to under actuated in which single control input, acceleration of cart, controlled the angle and position of cart. This ...