Unmanned Boat Recognition and Grasp
Project Type: Robotics / Software / Hardware
Role: Hardware & Software designer, Hardware & Software Engineer.
This is a research project that I’ve done in Robotics and Intelligence Manufacturing Laboratory at the Chinese University of Hong Kong, Shenzhen. This project implements an unmanned boat carrying a robot arm and a depth camera. The boat was able to autonomously recognize targets (such as a moving item) based on the features of the target and tried to catch the target.
My Contribution
I cooperated with my teammates to design the project framework and process. We composed a scenario that used an unmanned boat to recognize and collect trash floating on the sea. My job was to control the robot arm that made use of forward and inverse Kinematics. Since the target was moving, the grasping required accuracy and rapidity. I decided to adjust the orientation of the robot arm at each time frame based on the variance between the location of the robot arm and the location of the moving object. When the variance was small enough, we were confident that the robot could catch the target. I also cowork on the predicting algorithm that predicts the moving pattern of the target so as to increase the success rate of grasping. To implement this algorithm, we created a buffer to store the previous locations and predict the speed and orientation of the motion.
Meaning to Me
This is my first project on Robotics and there is no double that it triggered my interest in Robotics. When designing the theme of the project, I started to think about how robotics and artificial intelligence can be beneficial to our life. And I eventually got my answer - in most cases, robots can do the jobs that are hard to be completed by human beings due to the limitation of our nature. As a result, I view robotic development as an unparalleled feat that breaks through the limitation of human beings. And this belief drove me to be a robotic engineer. Apart from that, working with capable teammates also stimulated me to learn more and work harder to become a competent engineer.
Resources
Github: Click here to view