“Techno-Social Excellence” Marathwada Mitra Mandal’s Institute of Technology Lohgaon, Pune-47 Department of Mechatronics Engineering B.E. Project Stage I On AUTONOMOUS MOBILE ROBOT By 1. BMX29 VRUSHABH JAIN 2. BMX30 ABHISHEK GAIKWAD 3. BMX35 ANAND BARAPATRE 4. BMX07 BHAGYASHRI GADHE Under the Guidance of PROF. S. V .GOLANDE MMIT Lohgaon B.E. Project Stage I AY 2023-24 1 CONTENTS 1. INTRODUCTION 2. LITERATURE REVIEW 3. PROBLEM DEFINITION 4. OBJECTIVES 5. METHODOLOGY 6. CAD MODEL 7. SIMULATION/ANALYSIS 8. REFERENCES MMIT Lohgaon B.E. Project Stage I AY 2023-24 2 INTRODUCTION Autonomous Mobile Robots are machine which are capable of : • Self-contained machines. • Equipped with sensors. • Capable of perceiving their environment. • Make autonomous decisions based on sensor input. • Execute movements and actions independently. • Operate without human intervention. MMIT Lohgaon B.E. Project Stage I AY 2023-24 3 LITERATURE REVIEW Sr.No Title of Papers Author Journal /conference name with year of publication 1 Navigation of Autonomous Mobile Robots in Unstructured Environments John A. Smith, Mary L. Johnson International Conference on Robotics and Automation (ICRA) 2019 2 A systematic review on recent advances in autonomous mobile robot navigation Chen, Y., Wang H., Liu, J Robotics and ComputerIntegrated Manufacturing (2023) 3 A novel approach to autonomous mobile robot navigation using reinforcement learning Liu, Y., Wang, Z., & Zhang, Y. IEEE Transaction on Robotics (2023) MMIT Lohgaon B.E. Project Stage I Description This research paper presents a comprehensive study on the navigation of autonomous mobile robots operating in unstructured environments. The authors address the significant challenges associated with enabling robots to autonomously navigate through environments that lack predefined paths or have dynamic obstacles. This paper presents a systematic review on recent advances in autonomous mobile robot navigation. The authors focus on path planning algorithms, which are essential for enabling robots to navigate safely and efficiently in complex environments. They review a variety of path planning algorithms, including traditional algorithms such as A* and Dijkstra's algorithm. This paper presents a novel approach to autonomous mobile robot navigation using reinforcement learning. They proposed approach does not require any prior knowledge of the environment or the robot's dynamics. Instead, the robot learns to navigate by trial and error, using reinforcement learning to reward itself for reaching its goal and penalize itself for colliding with obstacles. AY 2023-24 4 PROBLEM DEFINITION Design and development of an Autonomous mobile robot for real-world applications Challenges with autonomous mobile robots: •Executing specific tasks autonomously, such as delivery, inspection, or manipulation •Navigating in complex, dynamic, and unstructured environments while avoiding obstacles. •Efficiently planning collision-free paths, especially in real-time •Accurately determining the robot's position within a map of its environment (localization). MMIT Lohgaon B.E. Project Stage I AY 2023-24 5 OBJECTIVES • Navigation: One of the primary objectives of autonomous mobile robots is to navigate their environment safely and efficiently. This involves obstacle avoidance, path planning, and localization. • Task Execution: Autonomous mobile robots are often designed to perform specific tasks, such as delivering items, cleaning floors, or inspecting infrastructure. MMIT Lohgaon B.E. Project Stage I AY 2023-24 6 METHODOLOGY Creation of the Robot in CAD Selection of Hardware Components Creating the Robot for Simulation (Gazebo) URDF FILE Developing a ROS package to control Motors Integrating Navigation Stack Calibration and Testing MMIT Lohgaon B.E. Project Stage I Chassis Lidar, IMU sensor Encoders Motors, Wheels Motor driver Raspberry Pi Battery Implement algorithms localization, and mapping (SLAM). Path Planning: Creating algorithms for route planning and obstacle avoidance. AY 2023-24 7 CAD Model (Rough) MMIT Lohgaon B.E. Project Stage I AY 2023-24 8 SIMULATION/ ANALYSIS ROS and Gazebo Integration: ✓ Developed ROS packages for encapsulating robot functionalities. ✓ Utilized ROS nodes for sensor data processing, motion planning, and communication. ✓ Integrated Gazebo as the simulation platform for a realistic 3D environment. ✓ Used Gazebo plugins to synchronize ROS nodes with the simulated robot model Simulation Results: ➢ Validated behavioral algorithms, including obstacle avoidance and path planning. ➢ Assessed sensor performance, especially LiDAR and various conditions. MMIT Lohgaon B.E. Project Stage I AY 2023-24 9 SIMULATION Mapping: ✓ Implemented mapping algorithms to generate accurate representations of the environment. ✓ Utilized SLAM techniques to create maps while the robot explored the simulated space. Fig - Visualization of Mapping results during simulation MMIT Lohgaon B.E. Project Stage I AY 2023-24 10 Guessing Robot's estimated pose using AMCL SIMULATION Localization: ✓ Developed and tested localization algorithms to accurately position the robot within the simulated environment. Fig - Visualization of localization results during simulation MMIT Lohgaon B.E. Project Stage I AY 2023-24 11 SIMULATION Navigation: ✓ Implemented navigation algorithms to plan and execute the robot's movements. ✓ Validated the robot's ability to follow predefined paths and adapt to dynamic changes in the environment. Fig - Visualization of Navigation results during simulation MMIT Lohgaon B.E. Project Stage I AY 2023-24 12 STAGE I WORK: In the initial phase of our Autonomous Mobile Robot project, we've focused on foundational aspects such as Preliminary Design: ✓ Developed a basic CAD model to visualize the structural layout Defined Hardware specifications: ✓ Collaboratively outlined the robot's sensory requirements, considering accuracy and range criteria. ✓ Determined power system specifications, including anticipated voltage and battery type for seamless integration. ROS and Gazebo Integration: ✓ Utilized ROS and gazebo for simulation and experimentation purposes. Budget Allocation: ✓ Allocated a preliminary budget for materials and components. MMIT Lohgaon B.E. Project Stage I AY 2023-24 13 REFERENCES [1] [ROS Topics]," in IEEE Robotics & Automation Magazine, vol. 17, no. 1, pp. 13-14, March 2010, doi: 10.1109/MRA.2010.935808. [2] N. DeMarinis, S. Tellex, V. P. Kemerlis, G. Konidaris and R. Fonseca, "Scanning the Internet for ROS: A View of Security in Robotics Research," 2019 International Conference on Robotics and Automation (ICRA), Montreal, QC, Canada, 2019, pp. 8514-8521, doi: 10.1109/ICRA.2019.8794451. [3] Bruno Steux and Oussama El Hamzaoui. tinySLAM: A SLAM algorithm in less than 200 lines C-language program. In ICARCV, pages 1975--1979. IEEE, 2010. [4] M. Köseoğlu, O. M. Çelik and Ö. Pektaş, "Design of an autonomous mobile robot based on ROS," 2017 International Artificial Intelligence and Data Processing Symposium (IDAP), Malatya, 2017, pp. 1-5, doi: 10.1109/IDAP.2017.8090199 MMIT Lohgaon B.E. Project Stage I AY 2023-24 14