IFAC PapersOnLine 50-1 (2017) 12759–12764 ScienceDirect ScienceDirect Available online at www.sciencedirect.com 2405-8963 © 2017, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved. Peer review under responsibility of International Federation of Automatic Control. 10.1016/j.ifacol.2017.08.1830 © 2017, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved. Keywords: Area coverage, unicycle robot, pattern generation, autonomous systems. 1. INTRODUCTION Nature is inundated with exquisite patterns occurring in various places, varied forms and different sizes. To list a few, patterns can be seen in galactic paths, spiral aerial combat paths’ of dragon flies, etc. Deriving motivation from these recurring patterns, the research community on robotics and control has actively pursued the generation of the patterns. Depending on the shapes and types of pat- terns, they can find several applications like exploration, monitoring, and so on. Researchers in the robotics and control community have widely explored the area of pattern generation over the past few decades. The patterns generated using au- tonomous agents can be broadly classified in two ways: • inter-agent spacial formations, • single/multiple agent(s) trajectories. Inter-agent spacial formations: As the name suggests, mul- tiple autonomous agents are required to generate patterns created out of the spacial configurations of these agents. This problem is commonly referred to as ‘formation con- trol’ wherein the control laws ensure desired inter-agent spacings like circles, triangles, etc. The formations could be generated in two dimensional spaces or three. They can also be either static in space or dynamic. Single/multiple agent(s) trajectories: In this cases, pat- terns are traced out by the trajectories of autonomous agents. This problem has been addressed using both single and multiple agents in both two and three dimensions. In this work, patterns are generated by an autonomous agent’s trajectories by suitably designing the control law -4.5 24.3 Fig. 1. Patterns Based on Relative Heading Based Control Law as shown in Fig. 1. So, present framework of the paper has close association with the work presented in Galloway et al. - Becker et al.. Galloway et al., explore the cyclic pursuit scheme in which each agent employs a constant bearing (CB) pursuit law, where the formation shape is defined by control law parameters. In his other work Galloway et al. present a modified version of CB pursuit law which achieves a fixed formation around a target independent of the initial conditions. Tsiotras et al. present an exten- sion of the classical consensus algorithm for multi-agent systems to achieve consensus outside the convex hull of Abstract: The objective of this paper is to generate planar patterns using a unicycle based robot. These patterns are annular and centred around a fixed point in the plane, which is termed as center point. The center could be any landmark for various purposes of target monitoring,exploration etc. The patterns are defined in terms of maximum and minimum radial distance from the center. The paper proposes a control strategy based on relative heading of the robot with respect to center. Analysis has been performed to guarantee the generation of annular patterns by imposing conditions over initial conditions and control law parameters. Verification of the theoretical results has been performed by means of simulation. Further, the control strategy has been implemented on a mobile robot to validate the results. * Shashank Agarwal, Twinkle Tripathy, Arpita Sinha and Aseem Borkar are with Systems and Control Engineering Group at IIT Bombay, India, (e-mail: shashank@sc.iitb.ac.in, twinkle.tripathy@sc.iitb.ac.in, aseem@sc.iitb.ac.in and arpita.sinha@iitb.ac.in) Shashank Agarwal * Twinkle Tripathy * Aseem Borkar * Arpita Sinha * Relative Heading based Pattern Generation