Method for automated prevention of a collision
In a method for automated avoidance of a collision of a vehicle with an object in the surroundings of the vehicle, multiple vehicle paths are predicted and each one is weighted with a vehicle path probability, the vehicle surroundings are recorded with an imaging vehicle sensor, an object in the vehicle surroundings is captured, at least one object path in the vehicle surroundings is predicted and is weighted with an object path probability, one of the vehicle paths is tested for collision with the at least one object path and if a collision is possible, a collision probability with the at least one object path is calculated, a weighting criterion for an overall collision probability of the vehicle with the object is ascertained and a test is performed of whether the weighting criterion exceeds a threshold and if the threshold is exceeded a collision avoidance maneuver is triggered.
1. A method for automated avoidance of a collision of a vehicle ( 10 ) with an object in the surroundings of the vehicle ( 10 ) in which:
a) multiple vehicle paths ( 12 , 14 , 16 ) for the vehicle ( 10 ) are predicted based on driving dynamics data of the vehicle ( 10 ) and an error probability distribution of the driving dynamics data of the vehicle ( 10 ) and each is weighted with a vehicle path probability;
b) the vehicle surroundings are captured by an imaging vehicle sensor;
c) an object ( 20 ) is detected in the vehicle surroundings;
d) at least one object path ( 22 , 24 ) in the vehicle surroundings is predicted and is weighted with an object path probability,
e) one of the vehicle paths ( 12 , 14 , 16 ) is tested for collision with the at least one object path ( 22 , 24 ) and if a collision is possible, a collision probability with the at least one object path ( 22 , 24 ) is calculated,
f) a weighting criterion for an overall probability of collision of the vehicle ( 10 ) with the object ( 20 ) is ascertained and tested for whether the weighting criterion exceeds a threshold,
g) a collision avoidance maneuver is triggered when the threshold is exceeded.
2. The method according to claim 1 , wherein multiple object paths ( 22 , 24 ) are predicted, and each is weighted with an object path probability, wherein one of the vehicle paths ( 12 , 14 , 16 ) and one of the object paths ( 22 , 24 ) are tested in pairs for the possibility of collision and if a collision is possible that pair is weighted with a collision probability.
3. The method according to claim 2 , wherein the multiple object paths ( 22 , 24 ) are ascertained on the basis of object capture data and an error probability distribution of the object capture data.
4. The method according to claim 1 , wherein the collision probability is ascertained from the vehicle path probability and the object path probability.
5. The method according to claim 1 , wherein the weighting criterion is formed by summation of multiple collision probabilities.
6. The method according to claim 1 , wherein the weighting criterion is formed by a weighted summation of multiple collision probabilities.
7. The method according to claim 1 , wherein the weighting criterion is formed by summation of a predetermined number of selected collision probabilities.
8. The method according to claim 1 , wherein the vehicle path probabilities are tested for exceeding a threshold and the collision probability is calculated only for vehicle paths ( 12 , 14 , 16 ) that exceed the threshold.
9. The method according to claim 1 , wherein multiple objects in the vehicle surroundings are captured, a priority is assigned to a respective object with respect to the other objects captured and method steps d through g are carried out for a predetermined number of highest priority objects.
10. Method according to claim 1 , wherein the collision avoidance maneuver comprises an audible, visible or tactile warning output by an automotive system.
11. Method according to claim 1 , wherein the collision avoidance maneuver comprises a speed of the vehicle ( 10 ) being altered by an automotive system.
12. Method according to claim 1 , wherein the collision avoidance maneuver comprises the vehicle being automatically guided along a path that does not intersect the at least one object path ( 22 , 24 ) and/or along a path of a lower collision probability.
13. Method according to claim 1 , wherein the collision avoidance maneuver is triggered by a microprocessor integrated into a camera module and is predetermined for at least one automotive system.
14. The method according to claim 1 , wherein one of the vehicle paths ( 14 ) is predicted based on the driving dynamics data of the vehicle ( 10 ) and the remaining vehicle paths ( 12 , 16 ) are predicted based on the error probability distribution of the driving dynamics data.
15. The method according to claim 1 , wherein the remaining vehicle paths are estimated from the error probability distribution of at least one of a yaw rate, a steering angle, a transverse acceleration, and a speed of the vehicle.
16. The method according to claim 1 , wherein one of the object paths ( 22 ) is predicted based on an instantaneous position and instantaneous velocity vector of the object and the remaining object path ( 24 ) is predicted based on an inaccuracy in the instantaneous position and the instantaneous velocity vector of the object.
17. The method according to claim 1 , wherein each of the vehicle paths and the at least one object path are tested in pairs for collision and given a collision probability and the overall probability of collision is obtained by adding the collision probabilities of each of the tested pairs of vehicle paths and the at least one object path.