Path planning apparatus and method for autonomous vehicle

1. A path planning apparatus for an autonomous vehicle, comprising:

a surrounding information detector configured to detect surrounding information around the vehicle;

a vehicle information detector configured to detect vehicle information regarding driving states of the vehicle;

a determiner configured to determine a lane keeping and a lane changing of the vehicle based on the surrounding information and the vehicle information; and

a driving path generator configured to generate a driving path and a velocity profile for a lane change using the surrounding information and the vehicle information when receiving a lane change signal from the determiner,

wherein the surrounding information includes road information regarding the vehicle, and the velocity, longitudinal acceleration, lateral acceleration, and position of a dynamic obstacle around the vehicle, and

wherein the driving path generator generates candidate paths within a current driving lane of the vehicle when a maneuver mode of the vehicle is a lane keeping mode, and the driving path generator generates candidate paths with respect to the current driving lane of the vehicle and a target lane for lane change respectively when the maneuver mode of the vehicle is a lane change mode, the candidate paths with respect to the current driving lane of the vehicle are used for the vehicle returning to an original driving lane when the lane change has failed, and

wherein the driving path generator includes a path selection module, the path selection module selects, as a local reference path, a candidate path that is closest to a target lane among the driving lane candidate paths, when the lane change has failed.

2. The path planning apparatus according to claim 1 , wherein the surrounding information detector is configured to detect the surrounding information using an imaging device, a distance sensor, and a position sensor.

3. The path planning apparatus according to claim 1 , wherein the vehicle information detector is configured to detect the velocity and position of the vehicle using a velocity sensor and a position sensor.

4. The path planning apparatus according to claim 1 , wherein the determiner is configured to bypass a signal generated in relation to a vehicle operator's manipulation to the driving path generator when a turn signal lamp is manipulated.

5. The path planning apparatus according to claim 1 , wherein the driving path generator include:

a candidate path generation module configured to generate driving lane candidate paths and target lane candidate paths;

a collision risk calculation module configured to calculate the degree of collision risk with respect to a dynamic obstacle positioned around the vehicle for each of the target lane candidate paths;

a velocity profile generation module configured to generate the velocity profile using the road information and the vehicle information; and

a path selection module configured to select the local reference path from among the target lane candidate paths based on the degree of collision risk or both the degree of collision risk and the velocity profile.

6. The path planning apparatus according to claim 5 , wherein the path selection module is configured to select, as the local reference path, a candidate path having a lowest degree of collision risk among the target lane candidate paths.

7. The path planning apparatus according to claim 5 , wherein the path selection module is configured to select, as the local reference path, a candidate path having a minimum cost through a cost function to which the degree of collision risk and the degree of proximity to a target path have been applied.

8. A path planning method for an autonomous vehicle, comprising:

receiving, by a controller, a lane change signal from a determiner;

generating, by the controller, driving lane candidate paths and target lane candidate paths based on the surrounding information and the vehicle information when the lane change signal is received, wherein the surrounding information includes road information regarding the vehicle, and the velocity, longitudinal acceleration, lateral acceleration, and position of a dynamic obstacle around the vehicle;

calculating, by the controller, the degree of collision risk with respect to the dynamic obstacle positioned around the vehicle for each of the target lane candidate paths;

determining, by the controller, whether there is a collision risk in all of the target lane candidate paths based on the calculated degree of collision risk;

generating, by the controller, a velocity profile of the vehicle based on the vehicle velocity and a target velocity when there is the collision risk in all of the target lane candidate paths; and selecting, by the controller, a local reference path from among the target lane candidate paths to which the velocity profile has been applied,

wherein the driving lane candidate paths are generated within a current driving lane of the vehicle when a maneuver mode of the vehicle is a lane keeping mode, and the driving land candidate paths and the target lane candidate path are generated with respect to the current driving lane of the vehicle and the target lane for lane change respectively when the maneuver mode of the vehicle is a lane change mode, the driving lane candidate paths are used for the vehicle returning to an original driving lane when the lane change has failed, and

wherein the local reference path is selected as a candidate path that is closest to a target lane among the driving lane candidate paths, when the lane change has failed.

9. The path planning method according to claim 8 , wherein the selecting of the local reference path includes:

calculating, by the controller, the degree of collision risk with respect to the dynamic obstacle for each of the target lane candidate paths to which the velocity profile has been applied;

determining, by the controller, whether there is the collision risk in all of the target lane candidate paths to which the velocity profile has been applied; and

selecting, by the controller, as the local reference path, a candidate path having a lowest degree of collision risk among the target lane candidate paths to which the velocity profile has been applied, when there is no collision risk in one or more candidate paths among the target lane candidate paths to which the velocity profile has been applied.

10. The path planning method according to claim 9 , further comprising:

returning, by the controller, the vehicle to the original driving lane when there is the collision risk in all of the target lane candidate paths to which the velocity profile has been applied, in the determining whether there is a collision risk in all of the target lane candidate paths to which the velocity profile has been applied.

11. The path planning method according to claim 10 , further comprising:

setting, by the controller, the target velocity by estimating a velocity of the dynamic obstacle preventing the lane change of the vehicle and adjusting the estimated velocity of the dynamic obstacle with a predetermined velocity control rate, after returning to the original driving lane.

12. The path planning method according to claim 8 , further comprising:

selecting, by the controller, any one of the target lane candidate paths as the local reference path in consideration of the degree of collision risk with the dynamic obstacle and the degree of proximity to a target path when there is no collision risk in all of the target lane candidate paths, after calculating the degree of collision risk with respect to the dynamic obstacle.