Laser scanner as a parking assistant
Fraunhofer ESK is responsible for the development of a driving assistance system which supports the driver of an electric vehicle (EV) in parking his car over a charging coil. Several sensors support the system. A laser scanner is mounted on the vehicle’s front bumper to detect the surrounding area within a given field of view in the horizontal plane. The result of the laser scanner acquisition is a point cloud which contains 2D points stored in an unorganized manner.
The point coordinates in the point cloud are useful to determine the position of the vehicle. In particular, the process of parking the vehicle by means of a laser scanner is articulated in two steps: initially, a reference map is calculated from the laser data acquired while driving in the vicinity of the charging coil. This step is executed once for each charging station. To compute the reference map, several laser scanner acquisitions (scans) should be combined into a common coordinate systems. Indeed, the laser scanner moves together with the vehicle and simultaneously generates scans that refer to different coordinate systems. By fixing the common coordinate system, the position of every measured point is uniquely determined. The reference map stores the positions of (at least three) key points and the position of the charging coil.
Once the reference map is available, the ego-position of the EV in relation to the charging coil can be calculated anytime. This is particularly useful since the coil is built into the road, therefore the driver cannot see it while driving over it. In this phase, the key points stored in the reference map are matched with corresponding points detected in real-time. The reference map can be either retrieved from the cloud or the infrastructure. It is expected that the positioning method delivers an accuracy of 5-10cm. A display shows the driver the current EV position in relation to the coil. This allows him to park over the coil, even if the road is not visible due to snow or other unfavorable weather conditions.