Papers by Senthilkumar Somasundaram
Single Robot Motion Planning using Fuzzy-genetic Technique
Control and Intelligent Systems
The autonomous multi-robot exploration and coverage is a well-addressed problem in the field of r... more The autonomous multi-robot exploration and coverage is a well-addressed problem in the field of robotics. Exploration and coverage is the task of guiding robots in such a way that they cover the environment in an efficient and effective manner. The Behaviour-based system and Role-based system for multi-robot control is popular in autonomous multirobot exploration, because they are robust to the dynamic interactions inherent in any multi-robot systems. In this paper, the existing Multi-Robot exploration and coverage systems under the characterization of behaviour-based and role-based are identified and their strengths and shortcomings discussed. Existing solutions to this problem differ primarily by the type of coordination that exists between the robots. The level of coordination depends on the type of communication the robots are expected to share.
Spanning tree based Terrain Coverage by multi-robots in unknown environments
ABSTRACT Terrain Coverage algorithms have been studied in the robotics literature and other navig... more ABSTRACT Terrain Coverage algorithms have been studied in the robotics literature and other navigation tasks because of their numerous applications including vacuum cleaning, lawn mowing, harvesting, mine clearing, inspection and intrusion detection. There are several remarkable research attempts on Terrain Coverage based on genetic algorithms, spanning trees, exact cell decomposition, and spiral filling paths. This paper presents a novel simultaneous on-line coverage strategy for multi robots, which is structured and assures complete and robust coverage of the surface regardless of the shape of the unknown environment. In particular, we study ant-robots and how they can cover terrain by leaving markings in the terrain, similar to what ants do. These markings can be sensed by all robots and allow them to cover the unknown terrain without direct communication with each other. A real-time heuristic search method is used to implement ant-robots and simulation results for simultaneous online terrain coverage are presented.
An efficient global optimization approach to multi robot path exploration problem using hybrid genetic algorithm
... Artificial Bee Colony Optimization Algorithm. Preetha Bhattacharjee, Pratyusha Rakshit, Goswa... more ... Artificial Bee Colony Optimization Algorithm. Preetha Bhattacharjee, Pratyusha Rakshit, Goswami Chakraborty, Amit Konar in Current (2011). 1 reader Save reference to library ยท Related research. A hybrid genetic algorithm and ...
An Autonomous Mobile Robot (AMR) is a machine able to extract information from its environment an... more An Autonomous Mobile Robot (AMR) is a machine able to extract information from its environment and use knowledge about its world to move safely in a meaningful and purposeful manner. Robot Navigation and Obstacle Avoidance are from the most important problems in mobile robots, especially in unknown environments. It must be able to interact with other objects safely. Several techniques such as Fuzzy logic, Reinforcement learning, Neural Networks and Genetic Algorithms, have applied to AMR in order to improve their performance. During the past several years Hybrid Genetic-fuzzy method has emerged as one of the most active and fruitful areas for research in the application of intelligent system design. The objective of this work is to provide a Hybrid method by which an improved set of rules governing the actions and behavior of a simple navigating and obstacle avoiding AMR. Genes are in the form of distances and angles labels. The chromosomes are represented as a rule written in a Boolean algebraic form. The method used to enhance the performance employs a simulation model designed by using Visual Basic software.
Multi-robot exploration and terrain coverage in an unknown environment
Robotics and Autonomous Systems, 2012
ABSTRACT Terrain exploration and coverage are required for a variety of applications such as mine... more ABSTRACT Terrain exploration and coverage are required for a variety of applications such as mine clearing, intrusion detection and other humanitarian missions like search and rescue operations, for example, fire or blast in a building. During an emergency situation within a building it is crucial to explore the area as fast as possible in order to search and find the wounded people and other hazards. On account of the prevailing breakdown of communication in indoor environments in some situations, it is suggested that the robots can communicate indirectly with the use of markings in the environment. The Spanning Tree Coverage (STC) method, proposed for this problem, suffers in environments that have partially occupied cells and narrow door openings between rooms. In this paper, we consider an extension of the Simultaneous Multiple STC (S-MSTC) algorithm, which we proposed in our previous work on multiple autonomous robots used in exploration and coverage in an unknown terrain. The proposed extended S-MSTC (ES-MSTC) uses ant-type robots to cover the terrain leaving marks on the terrain, which can be sensed by the robots and allow them to cover the terrain, similar to the nature of ants. This algorithm can handle partially occupied cells and narrow door openings in the terrain and performs a complete coverage of the surface regardless of the shape of the environment by constructing multiple spanning trees simultaneously. We present a simulation study and compare the performance of the ES-MSTC algorithm with other existing algorithms.
Machine Learning Algorithm on Assets' Behaviour During and Post Covid Pandemic
Proceedings of the World Multi-Conference on Systemics, Cybernetics and Informatics
Uploads
Papers by Senthilkumar Somasundaram