Method and system for fusing occupancy maps

1. A method, comprising:

obtaining a first occupancy map and a second occupancy map,

wherein:

each of the first occupancy map and the second occupancy map include a plurality of cells,

a position of each of the plurality of cells in the first occupancy map is the same as a position of a corresponding one of the plurality of cells in the second occupancy map,

the first occupancy map represents free and occupied space around a vehicle at a first time instance, and

the second occupancy map represents the free and occupied space around the vehicle at a second time instance;

determining at least one fusion parameter representing a potential dissimilarity between a respective one of the plurality of cells of the first occupancy map and the corresponding one of the plurality of cells of the second occupancy map;

determining a fused occupancy map representing the free and occupied space around the vehicle by fusing each of the plurality of cells of the first occupancy map with the corresponding one of the plurality of cells of the second occupancy map using a fusion rule, wherein the at least one fusion parameter controls an influence the respective one of the plurality of cells of the first occupancy map and/or the corresponding one of the plurality of cells of the second occupancy map has on the fused occupancy map; and

controlling at least one function of the vehicle based on the fused occupancy map.

2. The method of claim 1 , wherein determining the at least one fusion parameter is based on an estimated motion of the vehicle with respect to the first occupancy map and/or the second occupancy map.

3. The method of claim 2 , wherein determining the at least one fusion parameter comprises taking an expected error of the estimated motion of the vehicle into account in dependence on the estimated motion.

4. The method of claim 2 , wherein the estimated motion of the vehicle comprises an estimated linear velocity of the vehicle and/or an estimated yaw rate of the vehicle.

5. The method of claim 2 , wherein

determining the at least one fusion parameter is based on a plurality of predetermined calculation rules,

each of the calculation rules is associated with a predetermined condition of the estimated motion of the vehicle,

a respective one of the calculation rules is selected for determining the at least one fusion parameter, and

the respective calculation rule is selected with respect to the estimated motion of the vehicle meeting the predetermined condition associated with the respective calculation rule.

6. The method of claim 5 , wherein

the calculation rules comprise at least a first calculation rule and a second calculation rule,

the first calculation rule is associated with a first condition of the estimated motion in which the vehicle is assumed to move along a non-linear course, and

the second calculation rule is associated with a second condition of the estimated motion in which the vehicle is assumed to move along a linear course.

7. The method of claim 6 , wherein the calculation rules comprise a third calculation rule associated with a third condition of the estimated motion in which the vehicle is assumed not to move or to move with a velocity below a predetermined threshold.

8. The method of claim 5 , wherein at least one of the calculation rules comprises an offset term which is non-zero if the vehicle is assumed not to move or to move with a velocity below a predetermined threshold.

9. The method of claim 1 , wherein

the fusion rule is configured such that the influence depends on the potential dissimilarity represented by the at least one fusion parameter, and

the fused occupancy map represents the free and occupied space around the vehicle at a third time instance being closer or equal to the respective first or second time instance associated with the one of the first and second occupancy map.

10. The method of claim 1 , wherein

each of the plurality of cells is associated with a probabilistic value representing a ratio between a probability that the respective cell is occupied and a probability that the respective cell is free,

the fusion rule is adapted to determine the probabilistic values of the fused occupancy map per cell by fusing pairs of probabilistic values in dependence on the fusion parameters,

one member of a respective pair is the probabilistic value of respective one of the plurality of cells of the first occupancy map and another member of the respective pair is the probabilistic value of the corresponding one of the plurality of cells of the second occupancy map,

the respective one of the plurality of cells of the first occupancy map and the corresponding one of the plurality of cells share a positional property with respect to the space around the vehicle, and

a respective probabilistic value determined by fusing the probabilistic values of the respective pair is associated with a respective cell of the fused occupancy map sharing the positional property of the respective one of the plurality of cells of the first occupancy map and the corresponding one of the plurality of cells.

11. The method of claim 1 , comprising:

determining a raw sequence of the first occupancy maps and the second occupancy maps on the basis of consecutive sensor measurements, associated with consecutive time instances;

determining a filtered sequence of occupancy maps, wherein at least one member of the filtered sequence is the fused occupancy map determined by fusing two of the occupancy maps of said raw sequence.

12. A computer system for fusing occupancy maps, the computer system being configured to fuse occupancy maps by carrying out the method of claim 1 .

13. The computer system of claim 12 , comprising an input for receiving sensor measurements and an output for providing at least the fused occupancy map.

14. A non-transitory computer readable medium comprising instructions for carrying out the method of claim 1 .

15. The method of claim 1 , wherein

determining the at least one fusion parameter comprises determining a plurality of fusion parameters;

one of the plurality of fusion parameters is associated with a subset of the plurality of cells of the first occupancy map and corresponding ones of the plurality of cells of the second occupancy map; and

the subset of the plurality of cells of the first occupancy map and corresponding ones of the plurality of cells of the second occupancy map are at approximately the same distance from the vehicle.

16. A system for fusing a first occupancy map representing free and occupied space around a vehicle at a first time instance and a second occupancy map representing the free and occupied space around the vehicle at a second time instance, each of the first occupancy map and the second occupancy map including a plurality of cells, a position of each of the plurality of cells in the first occupancy map is the same as a position of a corresponding one of the plurality of cells in the second occupancy map, the system comprising:

a processor and

memory associated with the processor,

wherein the processor is configured to:

determine at least one fusion parameter representing a potential dissimilarity between a respective one of the plurality of cells of the first occupancy map and the corresponding one of the plurality of cells of the second occupancy map;

determine a fused occupancy map representing the free and occupied space around the vehicle by fusing each of the plurality of cells of the first occupancy map with the corresponding one of the plurality of cells of the second occupancy map using a fusion rule, wherein the at least one fusion parameter controls an influence the respective one of the plurality of cells of the first occupancy map and/or the corresponding one of the plurality of cells of the second occupancy map has on the fused occupancy map; and

control at least one function of the vehicle based on the fused occupancy map.

17. The system of claim 16 , wherein

the at least one fusion parameter is based on an estimated motion of the vehicle with respect to the first occupancy map and/or the second occupancy map,

the at least one fusion parameter is determined based on a selected one of a plurality of predetermined calculation rules,

each of the calculation rules is associated with a predetermined condition of the estimated motion of the vehicle, and

the selected one of the calculation rules is selected based on the estimated motion of the vehicle meeting the predetermined condition associated with the selected one of the calculation rules.

18. The system of claim 17 , wherein

the calculation rules comprise at least a first calculation rule and a second calculation rule,

the first calculation rule is associated with a first condition of the estimated motion in which the vehicle is assumed to move along a non-linear course, and

the second calculation rule is associated with a second condition of the estimated motion in which the vehicle is assumed to move along a linear course.

19. The system of claim 18 , wherein the calculation rules comprise a third calculation rule associated with a third condition of the estimated motion in which the vehicle is assumed not to move or to move with a velocity below a predetermined threshold.

20. The system of claim 17 , wherein at least one of the calculation rules comprises an offset term which is non-zero if the vehicle is assumed not to move or to move with a velocity below a predetermined threshold.