Publication: Direction sensing using robotic vision
Loading...
Date
2023-07
Authors
Tan, Yeong Ming
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
In the field of robotics, vision-based navigation is critical for autonomous robots to perceive their environment and make real-time decisions. This paper presents a direction sensing algorithm for robotic vision that utilizes Canny edge detection to estimate the position, and motion of obstacles approaching the camera's field of view. The algorithm continuously captures images and processes them using various image processing techniques such as blurring, noise filtering, and segmentation to identify significant blobs in the scene. These blobs represent desired path in the environment, allowing the robot to prioritize and navigate around them effectively. The proposed algorithm offers a robust, flexible, and efficient solution for obstacle detection and avoidance, ensuring safe and reliable navigation for autonomous robots in various environments and situations. The machine vision algorithm achieves a 100% success rate in detecting obstacles approaching from the left or right, while the success rate drops to 20% when the obstacle approaches from the center, and the algorithm achieves an 80% accuracy in obstacle avoidance for multiple obstacles.