Vol.:(0123456789) 1 3
Journal of Ambient Intelligence and Humanized Computing
https://doi.org/10.1007/s12652-020-02566-y
ORIGINAL RESEARCH
Improved ant colony optimization for achieving self‑balancing
and position control for balancer systems
Rupam Singh
1
· Bharat Bhushan
1
Received: 11 April 2020 / Accepted: 18 September 2020
© Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract
The balancer systems represent feedback in loop-based underactuated system which is electromechanical, multivariate, and
nonlinear. This paper develops a self-balancing controller using an improved ant colony optimization (ACO) to optimize the
proportional integral derivative controller (PID) controller. The proposed controller achieves self-balancing control for a ball
on the plate by controlling the plate inclination angle. Initially, the modelling of the ball balancer system is achieved with the
help of a two degree of freedom (2DoF) ball balancer system controlled by a PID controller. Further, ACO is employed to
autonomously evaluate the condition of a process and fnd the optimal tuning parameters for the PID controller. The transition
probability of the ACO is revised to improve the response and convergence speed of the algorithm resulting in an improved
ACO. The developed control schemes were applied with the 2DoF ball balancer model both in simulation as well as for the
real-time operation. The results depicted the performance of the proposed control scheme by analysing the characteristics
such as transient response and steady-state error. Further, stability analysis has been done for the developed control schemes
using describing function method for multiple frequencies. The results depicted the superiority of the improved ACO based
PID controller over the conventional PID controller.
Keywords Self-balancing control · Ball balancer setup · Proportional integral derivative control · Ant colony optimization
1 Introduction
The approximation of underactuated nonlinear systems
through automatic decision development and intelligent con-
trol methods (Murray et al. 2003) is an issue that appears in
many problems (Nelles 2001) and can be tackled through
various approaches. The diversity and complexity of these
systems have led researchers in the feld to analyse the action
of various linear, nonlinear, model-free, passivity, and intel-
ligent controllers. These controllers are focussed at achiev-
ing self-balancing control and steady-state operation for
various systems like inverted pendulum (Boubaker 2012;
Moness et al. 2020), the twin rotor multi input multi output
system (TRMS) (Chalupa et al. 2015), the ball beam system
(Andreev et al. 2002; Nowopolski 2013), hovercraft (Aranda
et al. 2006), Furuta pendulum (Acosta 2010), and ball and
plate system (Awtar et al. 2002). In general, linear control-
lers ofer a simple way of designing closed-loop control for
these systems. Various linear controllers were available in
the literature to solve the problems in underactuated systems
(de Jager 1998; Olivares and Albertos 2013; Aguilar-Avelar
and Moreno-Valenzuela 2015). But the complicated nonlin-
ear dynamics of underactuated systems afect the capability
of providing a plausible solution and limits the generalized
applications of control laws. Besides, the linearization of
the nonlinear systems is also carried which heavily afect
the speed of system response. This motivated the develop-
ment of several nonlinear control techniques to deal with the
problems in underactuated systems. Lots of nonlinear con-
trollers like Lagrangian, lambda method, and backstepping
controller (Chang 2008; Choukchou-Braham et al. 2014;
Rudra et al. 2017), have been evolved in the last few years.
These controllers faced some drawbacks during the early
stage of their development with the limitations of lambda
method in dealing with external load and back stepping con-
trollers lagging due to additional feedback. Apart from the
above controllers, the adaptive iterative learning methods
* Rupam Singh
singhrupam99@gmail.com
Bharat Bhushan
bharat@dce.ac.in
1
Department of Electrical Engineering, Delhi Technological
University, Shahbad Daulatpur, Delhi 110042, India