Laser sensor-based map generation

1. A laser sensor-based map generation method, comprising:

acquiring image data acquired by a visual sensor;

determining first point cloud data belonging to glass-like region in laser data based on the image data, the laser data being acquired by a laser sensor, the laser data and the image data corresponding to one same region, a time period in which the visual sensor acquires the image data being the same as a time period in which the laser sensor acquires the laser data;

adjusting a weight of the laser data according to the first point cloud data; and

fusing the first point cloud data and second point cloud data belonging to non-glass-like region in the laser data based on the adjusted weight of the laser data, to generate a map; and, wherein determining first point cloud data in laser data based on the image data comprises:

identifying a glass-like region in the image data; and

determining the first point cloud data in the laser data according to the glass-like region in the image data and a pre-received extrinsic parameter between the visual sensor and the laser sensor.

2. The method according to claim 1 , wherein adjusting a weight of the laser data according to the first point cloud data comprises any one or more of the following:

decreasing a first weight corresponding to the first point cloud data; or

increasing a second weight corresponding to the second point cloud data.

3. The method according to claim 2 , wherein decreasing the first weight corresponding to the first point cloud data comprises:

decreasing a confidence level of the first point cloud data; and

decreasing the first weight according to the decreased confidence level.

4. The method according to claim 2 , wherein increasing the second weight corresponding to the second point cloud data comprises:

increasing a confidence level of the second point cloud data; and

increasing the second weight according to the increased confidence level.

5. The method according to claim 1 , wherein the extrinsic parameter indicates a spatial rotation and translation relationship between a first coordinate system in which the visual sensor is located and a second coordinate system in which the laser sensor is located.

6. The method according to claim 1 , wherein identifying the glass-like region in the image data comprises:

segmenting an image corresponding to the image data into a plurality of subimages;

determining, according to one recognition model pre-trained or a plurality of recognition models pre-trained, whether each of the subimages belongs to glass-like region; and

determining the glass-like region in the image data according to subimages belonging to glass-like region.

7. The method according to claim 6 , wherein determining, according to the recognition model pre-trained, whether each of the subimages belongs to glass-like region comprises:

inputting the subimage into the recognition model, to obtain a probability of determining, based on the recognition model, that the subimage belongs to glass-like region; and

when the probability of determining, based on the recognition model, that the subimage belongs to glass-like region is greater than a probability threshold corresponding to the recognition model, determining that the subimage belongs to glass-like region.

8. The method according to claim 6 , wherein determining, according to the plurality of recognition models pre-trained, whether each of subimages belongs to glass-like region comprises:

for each of the plurality of recognition models, inputting the subimage into the recognition model to obtain a probability of determining, based on the recognition model, that the subimage belongs to glass-like region; and

when for each of the plurality of recognition models, the probability of determining, based on the recognition model, that the subimage belongs to glass-like region is greater than a probability threshold corresponding to the recognition model, determining that the subimage belongs to glass-like region.

9. The method according to claim 6 , wherein determining the glass-like region in the image data according to subimages belonging to glass-like region comprises:

using a combination of the subimages belonging to glass-like region as the glass-like region in the image data.

10. The method according to claim 1 , wherein fusing the first point cloud data and second point cloud data based on the adjusted weight of the laser data, to generate the map comprises:

fusing the first point cloud data and the second point cloud data based on the adjusted weight of the laser data, to obtain an initial map; and

optimizing the initial map to generate the map.

11. The method according to claim 10 , wherein fusing the first point cloud data and second point cloud data based on the adjusted weight of the laser data comprises:

registering the first point cloud data, to generate registered first point cloud data;

registering the second point cloud data, to generate registered second point cloud data;

calculating a cost function based on a coordinate vector of the registered first point cloud data, a coordinate vector of the registered second point cloud data, a decreased first weight corresponding to the registered first point cloud data, an increased second weight corresponding to the registered second point cloud data, and an attitude parameter between registered point cloud data;

optimizing the attitude parameter by performing iterative operation on the cost function; and

fusing the registered first point cloud data and the registered second point cloud data based on the optimized attitude parameter.

12. A non-transitory computer-readable storage medium, storing a computer program, the computer program, when invoked by a processor, causing the processor to perform:

acquiring image data acquired by a visual sensor;

determining first point cloud data belonging to glass-like region in laser data based on the image data, the laser data being acquired by a laser sensor, the laser data and the image data corresponding to one same region, a time period in which the visual sensor acquires the image data being the same as a time period in which the laser sensor acquires the laser data;

adjusting a weight of the laser data according to the first point cloud data; and

fusing the first point cloud data and second point cloud data belonging to non-glass-like region in the laser data based on the adjusted weight of the laser data, to generate a map;

wherein determining first point cloud data in laser data based on the image data comprises:

identifying a glass-like region in the image data; and

determining the first point cloud data in the laser data according to the glass-like region in the image data and a pre-received extrinsic parameter between the visual sensor and the laser sensor.

13. A mobile device, comprising a visual sensor, a laser sensor, a processor, a memory, and a computer program stored in the memory and run on the processor, the processor, when executing the computer program, the processor is caused to perform:

acquiring image data acquired by a visual sensor;

determining first point cloud data belonging to glass-like region in laser data based on the image data, the laser data being acquired by a laser sensor, the laser data and the image data corresponding to one same region, a time period in which the visual sensor acquires the image data being the same as a time period in which the laser sensor acquires the laser data;

adjusting a weight of the laser data according to the first point cloud data; and

fusing the first point cloud data and second point cloud data belonging to non-glass-like region in the laser data based on the adjusted weight of the laser data, to generate a map;

wherein determining first point cloud data in laser data based on the image data comprises:

identifying a glass-like region in the image data; and

determining the first point cloud data in the laser data according to the glass-like region in the image data and a pre-received extrinsic parameter between the visual sensor and the laser sensor.