Path planning method and system using the same

1. A path planning method comprising:

inputting a road distance point-cloud map, a first camera road image, and a second camera road image of a target road, wherein the road distance point-cloud map is created according to actual coordinates of a plurality of distance data points of the target road;

calibrating and fusing the road distance point-cloud map and the first camera road image to generate a road camera point-cloud fusion map and determining road-line information of the second camera road image to generate a road-segmented map of the target road; fusing the road-segmented map and the road camera point-cloud fusion map according to their feature points, thereby obtaining a plurality of distance data of the road-segmented map according to the plurality of distance data points, and planning a front driving path for the target road according to the plurality of distance data; and,

performing a path operation on the target road to plan the driving path when the number of the plurality of distance data points is larger than a threshold and a distance between each of the plurality of distance data points and the target road is less than a radius range.

2. The path planning method according to claim 1 , wherein the step of planning the driving path comprises:

computing the plurality of distance data points based on a vector quantization algorithm to determine initial road data of the target road;

clustering the initial road data based on a clustering algorithm; and

using circle equations and polynomials to fit a path of the target road, thereby planning the driving path.

3. The path planning method according to claim 2 , wherein the vector quantization algorithm is a k-means algorithm.

4. The path planning method according to claim 2 , wherein the clustering algorithm is density-based spatial clustering of applications with noise (DBSCAN).

5. The path planning method according to claim 1 , wherein the road distance point-cloud map is a road lidar point-cloud map.

6. The path planning method according to claim 5 , wherein the step of calibrating and fusing the road distance point-cloud map and the first camera road image comprises:

obtaining internal parameters of a camera for capturing the first camera road image and external parameters of the actual coordinates of the road distance point-cloud map and superposing the road distance point-cloud map and the first camera road image according to relationships between the internal parameters and the external parameters.

7. The path planning method according to claim 1 , wherein the step of determining the road-line information of the second camera road image to generate the road-segmented map comprises:

performing model initialization on the second camera road image to generate a model-initializing image;

segmenting the model-initializing image to generate a road-predicting map; and

computing the road-predicting map based on a probabilistic graphical model to generate a lane-segmented map and a road line-segmented map and fusing the lane-segmented map and the road line-segmented map to generate the road-segmented map.

8. A path planning system comprising:

an image-capturing device configured to capture a first camera road image and a second camera road image of a target road;

a point-cloud map-retrieving device configured to retrieve a plurality of distance data points of the target road and create a road distance point-cloud map according to actual coordinates of the plurality of distance data points; and

a processing device, connected to the image-capturing device and the point-cloud map-retrieving device, comprising:

a point-cloud image-fusing module configured to receive, calibrate, and fuse the road distance point-cloud map and the first camera road image to generate a road camera point-cloud fusion map;

a road area-recognizing module configured to receive the second camera road image and determine road-line information of the second camera road image to generate a road-segmented map of the target road; and

a path-planning module connected to the point-cloud image-fusing module and the road area-recognizing module and configured to receive the road-segmented map and fuse the road-segmented map and the road camera point-cloud fusion map according to their feature points, thereby obtaining a plurality of distance data of the road-segmented map according to the plurality of distance data points, and the path-planning module is configured to plan a front driving path for the target road according to the plurality of distance data,

wherein the path-planning module performs a path operation on the target road to plan the driving path when the number of the plurality of distance data points is larger than a threshold and a distance between each of the plurality of distance data points and the target road is less than a radius range.

9. The path planning system according to claim 8 , wherein the path-planning module is configured to:

compute the plurality of distance data points based on a vector quantization algorithm to determine initial road data of the target road;

cluster the initial road data based on a clustering algorithm; and

use circle equations and polynomials to fit a path of the target road, thereby planning the driving path.

10. The path planning system according to claim 9 , wherein the vector quantization algorithm is a k-means algorithm.

11. The path planning system according to claim 9 , wherein the clustering algorithm is density-based spatial clustering of applications with noise (DBSCAN).

12. The path planning system according to claim 8 , wherein the road distance point-cloud map is a road lidar point-cloud map.

13. The path planning system according to claim 8 , wherein the point-cloud image-fusing module is configured to obtain internal parameters of a camera for capturing the first camera road image and external parameters of the actual coordinates of the road distance point-cloud map and superpose the road distance point-cloud map and the first camera road image according to relationships between the internal parameters and the external parameters.

14. The path planning system according to claim 8 , wherein the road area-recognizing module is configured to:

perform model initialization on the second camera road image to generate a model-initializing image;

segment the model-initializing image to generate a road-predicting map; and

compute the road-predicting map based on a probabilistic graphical model to generate a lane-segmented map and a road line-segmented map and fusing the lane-segmented map and the road line-segmented map to generate the road-segmented map.

15. The path planning system according to claim 8 , wherein the image-capturing device is a camera.

16. The path planning system according to claim 8 , wherein the point-cloud map-retrieving device is an optical lidar.