The first simulation I ever did with the software: A ball falling on a ramp.
Note: there are invisible walls around the map to avoid objects from escaping.
A ball falling on a ramp and hitting a stack of cuboids.
My first simulated robot consisting of a sliding cuboid. The robot is driven by a force and a torque actring on its center of mass and tries to follow the ball. The desired rotational and translational speed is calculated using fuzzy logic while the corresponding force and torque is controled by a PID-control each.
Same scenario as in BallKicker1 but with an improved look of the robot. But the geometry used for collision detection is the same so that the hull and the wheels can move inside the wall or the ball.
Two robots of the kind of BallKicker2 following the ball at the same time.
Improved construction of the previous robots: it has now two wheels at the side and a ball to keep balance. The control and drive is the same.
The robot of BallKicker4 has received a cylindric transparent hull. This is also used for collision detection so the robot no more gets stuck at the wall that easily.
Same construction as in BallKicker5 but now there are torques applied between the wheels and the bodies. The torques are calculated by a PID-Control each. Due to the not very well suited PID-parameters the robot's behavior is worse than that of BallKicker4.
Robot construction of BallKicker5 but with a contact-, a range- and a camera-sensor. Those have no effect on the generated movie though.
Two robots of the kind of BallKicker5 but with a communication module each connected to each other. After the robots have "diced" who will start "catching" the ball the winner will follow it until it has touched it. Than it will notify th eother robot to do the same and stop moving - only adapting the orientation - until it is notified itself.
Construction from BallKicker5 but the balancing ball has been replaced by an idler gear. The camera has been moved while recording the movie and it's running with lower speed.