Radar reference map generation
Methods and systems are described that enable radar reference map generation. A high-definition (HD) map is received and one or more HD map objects within the HD map are determined. Attributes of the respective HD map objects are determined, and, for each HD map object, one or more occupancy cells of a radar occupancy grid are indicated as occupied space based on the attributes of the respective HD map object. By doing so, a radar reference map may be generated without a vehicle traversing through an area corresponding to the radar reference map.
1. A method comprising:
receiving or generating a high-definition (HD) radar reference map based on data collected from radar sensors of a radar system of a vehicle, wherein the HD radar reference map is generated based on a radar occupancy grid, and wherein the radar occupancy grid is generated based on radar signals transmitted from and received at the vehicle;
determining one or more HD map objects to be within the HD radar reference map, the one or more HD map objects being objects detected using the radar system;
determining attributes of the one or more HD map objects;
based on the attributes, forming a shape for each of the one or more HD map objects, the shape being a geometrical representation of that HD map object;
determining whether the shape of one of the one or more HD map objects is longer than a cell of the radar occupancy grid;
for each of the HD map object, indicating one or more cells of the radar occupancy grid as occupied space based on the shape of the one or more HD map objects occupying the cells;
sampling the shape of one of the one or more HD map objects comprising sampling at a first rate the shape of one of the one or more HD map objects in response to the shape of the one of the one or more HD map objects not being longer than one of the cells, and
oversampling at a second rate the shape of the one of the one or more HD map objects in response to the shape being longer than one of the cells, the second rate being greater than the first rate;
updating the HD radar reference map based on results of the sampling;
and autonomously controlling operation of the vehicle based on the updated HD radar reference map.
2. The method of claim 1 , wherein the HD radar reference map comprises statistical representations of the cells of the radar occupancy grid that are indicated as occupied space.
3. The method of claim 1 , wherein the attributes comprise one or more of a type, location, dimensions, orientation, linkage to a corresponding road, or radar hardware information for the one or more HD map objects.
4. The method of claim 1 , further comprising:
adjusting the shape of the one of the one or more HD map objects based on attributes of the one of the one or more HD map objects; and
indicating occupancy of the cells based on the adjusted shape.
5. The method of claim 4 , further comprising selecting a model for the one or more HD map objects from a plurality of models corresponding to respective types or attributes of the HD map objects,
wherein the adjustment is based on the model.
6. The method of claim 5 , further comprising:
inputting the attributes of the one or more HD map objects into the model; and
receiving an output from the model,
wherein the adjustment is based on the output from the model.
7. A method comprising:
receiving or generating a high-definition (HD) radar reference map based on data collected from radar sensors of a radar system of a vehicle;
determining one or more HD map objects to be within the HD radar reference map;
determining attributes of the one or more HD map objects, the one or more HD map objects being objects detected using the radar system;
for each of the one or more HD map objects, generating a Gaussian multivariate distribution for each cell of the HD radar reference map based on the attributes of the respective one or more HD map objects, wherein the generation of Gaussian multivariate distributions of the cells of the HD radar reference map includes generation of mean and covariance values for each of the cells based on the attributes, wherein the Gaussian multivariate distributions comprise metadata, wherein the metadata comprises at least one of i) information regarding shapes and dimensions of clusters of the Gaussian multivariate distributions, and ii) object associations relating the Gaussian variate distributions to certain objects, and
wherein the HD radar reference map is generated based on the Gaussian multivariate distributions; and
autonomously controlling operation of the vehicle based on the HD radar reference map.
8. A system comprising:
at least one processor; and
at least one computer-readable storage medium comprising instructions that, when executed by the processor, cause the system to receive or generate a high-definition (HD) radar reference map based on data collected from radar sensors of a radar system of a vehicle,
wherein the HD radar reference map is generated based on a radar occupancy grid, and wherein the radar occupancy grid is generated based on radar signals transmitted from and received at the vehicle,
determine one or more HD map objects to be within the HD radar reference map, the one or more HD map objects being objects detected using the radar system,
determine attributes of the one or more HD map objects,
based on the attributes, forming a shape for each of the one or more HD map objects, the shape being a geometrical representation of that HD map object,
determining whether the shape of one of the one or more HD map objects is longer than a cell of the radar occupancy grid,
for each of the HD map object, indicate one or more cells of the radar occupancy grid as occupied space based on the shape of the one or more HD map objects occupying the cells,
sampling the shape of one of the one or more HD map objects comprising sampling at a first rate the shape of one of the one or more HD map objects in response to the shape of the one of the one or more HD map objects not being longer than one of the cells, and
oversampling at a second rate the shape of the one of the one or more HD map objects in response to the shape being longer than one of the cells, the second rate being greater than the first rate,
updating the HD radar reference map based on results of the sampling, and autonomously controlling operation of the vehicle based on the updated HD radar reference map.
9. The system of claim 8 , wherein the HD radar reference map comprises statistical representations of the cells of the radar occupancy grid that are indicated as occupied space.
10. The system of claim 8 , wherein the attributes comprise one or more of a type, location, dimensions, orientation, or linkage to a corresponding road for the one or more HD map objects.
11. The system of claim 8 , wherein the instructions further cause the system to:
adjust the shape of the one of the one or more HD map objects based on attributes of the one of the one or more HD map objects to form an adjusted shape; and
indicate occupancy of the cells based on the adjusted shape.
12. The system of claim 11 , wherein:
the instructions further cause the system to select a model for the one or more HD map objects; and
the adjustment is based on the model.
13. The system of claim 12 , wherein:
the instructions further cause the system to
input the attributes of the one or more HD map objects into the model, and
receive an output from the model; and
the adjustment is based on the output from the model.
14. The method of claim 1 , wherein sampling the shape of the one of the one or more HD map objects comprises refraining from oversampling the shape of one of the one or more HD map objects in response to the shape of the one of the one or more HD map objects not being longer than one of the cells.
15. The method of claim 1 , wherein sampling the shape of the one of the one or more HD map objects comprises oversampling the shape of the of one of the one or more HD map objects in response to the shape being longer than the one of the one or more HD map objects.