Student presentation "Humanoid Robotic Systems"
As usual at the end of the semester, the students of the classes demonstrated their project work. In the lecture "Humanoid Robotic Systems" they worked with the Nao robot. We congratulate the student groups, who succeeded this semester:
Team “IndiaNao Jones”
This team developed a system enabling the NAO robot to search for a coloured object while walking through a “maze” described by black lines on a white ground. The NAO followed the lines and, once it has spotted the target object on a small box, NAO walked to it, picked it up, and walked and handed it over to a person. This project involved line detection and tracking, walking, face and voice recognition, object detection, and grasping.
This team made the NAO play the popular “battleship” board game. A grid board was created with foam “ships” on it, positioned in front of the NAO. The robot recognized the board with the current positioning of the ships and could play the game with a person by voice commands. The system contained a robust visual recognition of the board, end-effector control for placing the ships on the board, voice control and recognition, and the game logic.
Team “Do it NAO”
This team created a system which endowed the NAO with an object search, pick and throw skill. The NAO looked for a tiny ball in its surroundings, walked to it, and picked it up from the ground. It then detected the target location to throw the ball on. It oriented itself towards the target and it then threw the ball on it. This project contained a vision system for searching and tracking target objects, walking, and a control system for grasping as well as throwing.
Team “Nao The Constructor”
This team programmed the NAO to recognize a 2D geometric shape (e.g. a triangle, a circle, etc.) drawn on a paper and to reconstruct that particular shape by placing tiny foam boxes on a playground. For this task, the robot first recognized the drawn shape shown to him and grasped a foam box handed over by a person. The robot then determined the placement position on the playground, walked to that position, and placed the foam box there. Then the next box was given to the robot and it repeated the behaviour until it reconstructed the geometric shape by the boxes on the ground. Major parts of the system were visual shape recognition, odometry, map creation, walking, and position and orientation control.