The gait of most bipedal robots is controlled by precomputed trajectories, however, in robot soccer, a dynamic environment forces the robot to adapt their walking direction, speed and rotation to the changes . Our goal is capsulate the biped robot into an omnidirectional moving platform in the view of the mounted camera on head, and making gait parameterized with 3 pa-rameters: offset in forward and sidle direction and another rotation direction around Z axis. Several walking strategies have been developed, most of which are based on the Three-Dimensional Linear Inverted Pendulum . Firstly, foot trajectory is directly deduced from the foot planner from the gait command. Second, the center of pressure trajectory is defined based on ZMP discipline. COM trajectory is simply related to that of COP assuming the robot as a three-dimensional linear inverted pendulum. Third, inverse kinematics generates joint trajectories based on the former foot and COM trajectories.
Attached file →The software structure now is based on ROS framework, which greatly increases the efficiency of multitasking coordination and simplifies the process of application of new algorithm. There are four modules of the algorithm of our robot,vision is one part of them.Vision Module is used to recognize objects in the football court and convert them into information in geometric forms.The vision part mainly deal with two things: object recognition and distance measurement.We use one single camera,and our vision algorithm is based on computer vision(OpenCV).This year we use deep learning methods to increase the accuracy of the recognition of the ball and the sideline,and calculate the distance.Then the data will be used to decide the robot’s behavior.
Attached file →Self-localization is a state estimation problem. We choose the widely used particle filter algorithm to solve this problem. The robot’s camera gets the photos of the environment, the robot detects and gets the features in the photos by its algorithm. Then the robot computes the differences between features in the photos and features predicted. Through the differences, the robot gets its position and gesture. Also we make borders of the field be landmarks, our robot can get distance information by matching landmarks. If the center circle is found, our robots will locate based on the information the center circle. It is not difficult to compute the distance between the robot and the center the field by using some optical knowledge and geometry skills. Then combined with the magnetic location which can get the information of the direction of robots, we can know the exact position of our robots.
Attached file →The framework of the algorithm is composed of several super-states and each super-state is constructed by a group of subordinate states. In our case, those super-states include attack, listening to controller, standing up and defense. Correspondingly, those subordinate states, which are all basic actions able to be executed directly by the robot, involve searching the ball, approaching the ball, shoot at the goal and so on. Cooperation between robots is implemented on our robots. Our robots can share information of their position, where the ball and the goal is through WLAN, so that they can cooperate to get the ball into the goal. Also, our robots share what they are doing now, if they are doing the same thing, one robot will stop doing this for not obstructing the other.
Attached file →