This thesis documents a simulation-based investigation into the use of vision for the navigation of mobile robots. We developed characteristics such as goal-seeking, obstacle, and collision avoidance with visual data. In this thesis, we address the ability of autonomous robots to avoid conflicting circumstances, thereby allowing the robot to attain independent and goaloriented navigation. Current robotics research and development are primarily based on making the robotic system more autonomous and versatile. Tracking the objects is an extremely critical issue in computer vision, so we use techniques such as Canny edge detection, perspective transform along with libraries like OpenCV and TensorFlow are used to process the image. The processed images are used by different algorithms to perform lane following and traffic sign detection. The issue of moving a robot through severe conditions has pulled into much consideration. A robot may encounter obstacles of all structures, which is needed to be avoided intelligently. Therefore, algorithms that consider absolute path length and security are created by using a well-known A* path planning approach. The robot uses the A* path planning to create a cost map to select the most efficient path with minimum distance and risk. To run the simulation to visualize the robot's path planning, lane following, and traffic light detection, we built an environment and mapped it using the SLAM (Simultaneous Localization and Mapping) technique. The data obtained by performing all the algorithms are stored in the ROS bag. These data are then plotted to understand the behavior of the robot under different conditions


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Vision based obstacle avoidance and navigation system for mobile robots


    Beteiligte:

    Erscheinungsdatum :

    2020-12-01


    Medientyp :

    Sonstige


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Klassifikation :

    DDC:    629




    Mobile Robots with Dynamic Obstacle Avoidance

    Ali, Maram / Das, Saptarshi | IEEE | 2023


    Vision-Based Obstacle Avoidance of Wheeled Robots Using Fast Estimation

    Dippold, Amanda / Ma, Lili / Hovakimyan, Naira | AIAA | 2008


    Vision-Based Obstacle Avoidance of Wheeled Robots Using Fast Estimation

    Dippold, A. / Ma, L. / Hovakimyan, N. et al. | British Library Conference Proceedings | 2008


    Machine Vision-based Obstacle Avoidance for Mobile Robot

    Widodo, Nuryono Satya / Pamungkas, Anggit | BASE | 2019

    Freier Zugriff