Let the DLR Scout Rover Flip: Parameter Study Simulations

Abstract

The DLR Scout rover is symmetrical about the horizontal plane and can drive upside down equally well as upright. The rover was designed such, because its intended area of applications, planetary caves, is very difficult to access and navigate through, tipping over is just a matter of time. Entrance through skylights or drops inside the cave can cause the rover to end upside down with a high probability. There are however reasons to prefer being upright, mainly camera and payload, but also communication, although two antennas are planned for the final design. There is thus motivation for the rover to be able to flip itself, on its own or along an obstacle, in a controlled way. In order not to put the prototype at risk and for repeatability, a simulation parameter study is performed, with a few validation points using hardware. Results show that the rover is able to flip itself on a flat plane, when the wheels abruptly turn quickly in opposite directions on both rover sides. This maneuver is currently not possible on the prototype for safety reasons. The flip simulation along an obstacle shows that a simple inclined plane is enough, with higher success rate the faster the wheels turn. Other parameters like friction or the introduction of a ditch in front of the inclined plane, have little to no impact. Driving on the side of an inclined plane to induce a roll movement into the rover, mostly is unsuccessful in simulation (sliding or ending on one side) but mostly is successful with the prototype because imperfections are enough for the rover to end upside down if it comes to rest on a side.