Exhibiting Object Identification Through SLAM with Simultaneous Applications of ROS, YOLO, LIDAR, and Unitree GO1 Camera
Disciplines
Engineering | Robotics
Abstract (300 words maximum)
This research enhances the Unitree Go1 quadruped’s capabilities by integrating Simultaneous Localization and Mapping (SLAM) with You Only Look Once (YOLO) object detection. SLAM provides environmental mapping but lacks object recognition, while YOLO enables real-time detection of objects within the robot’s camera view. By combining SLAM, YOLO, Robot Operating System (ROS), and Light Detection and Ranging (LIDAR) with a Raspberry Pi 4, we aim to create a system that not only maps the robot’s surroundings but also accurately identifies and labels key objects. This integrated approach will be tested in a controlled environment to improve the Go1’s potential for diverse applications such as security, tracking, and delivery tasks.
Academic department under which the project should be listed
SPCEET - Robotics and Mechatronics Engineering
Primary Investigator (PI) Name
Muhammad Hassan Tanveer
Exhibiting Object Identification Through SLAM with Simultaneous Applications of ROS, YOLO, LIDAR, and Unitree GO1 Camera
This research enhances the Unitree Go1 quadruped’s capabilities by integrating Simultaneous Localization and Mapping (SLAM) with You Only Look Once (YOLO) object detection. SLAM provides environmental mapping but lacks object recognition, while YOLO enables real-time detection of objects within the robot’s camera view. By combining SLAM, YOLO, Robot Operating System (ROS), and Light Detection and Ranging (LIDAR) with a Raspberry Pi 4, we aim to create a system that not only maps the robot’s surroundings but also accurately identifies and labels key objects. This integrated approach will be tested in a controlled environment to improve the Go1’s potential for diverse applications such as security, tracking, and delivery tasks.