Method, apparatus and server for real-time learning of travelling strategy of driverless vehicle

1. A method for real-time learning of a travelling strategy of a driverless vehicle, comprising:

determining whether a first travelling strategy of the driverless vehicle is unable to be generated according to routing information depending on a high-precision map and surrounding obstacle information of the driverless vehicle;

in response to determining that the first travelling strategy of the driverless vehicle is unable to be generated according to the routing information depending on the high-precision map and the surrounding obstacle information, of the driverless vehicle, recording travelling trajectories of other vehicles on a road;

when a number of vehicles on a same travelling trajectory is greater than a preset first number threshold, generating a second travelling strategy using the same travelling trajectory; and

controlling the driverless vehicle to travel using the second travelling strategy,

wherein the method further comprises:

when a third travelling strategy is received by a second driverless vehicle from a server, determining whether the first travelling strategy is able to be generated by the second driverless vehicle; and,

in response to the first travelling strategy being unable to be generated by the second driverless vehicle, controlling the second driverless vehicle to travel using the third travelling strategy and transmitting a used report indicating that the third travelling strategy has been used to the server, such that the server records a number of used reports and a duration of receiving used reports, and updates the high-precision map in response to the number of used reports is greater than a report threshold or the duration is longer than a duration threshold.

2. The method according to claim 1 , further comprising:

reporting the second travelling strategy to the server, and/or transmitting the second travelling strategy to at least one other driverless vehicle.

3. The method according to claim 1 , further comprising:

controlling the second driverless vehicle to travel using the first travelling strategy and transmitting to the server a report indicating that the second driverless vehicle does not use the third travelling strategy, if the first travelling strategy is able to be generated by the second driverless vehicle.

4. A non-volatile computer-readable storage medium storing a computer program, wherein the computer program implements the method according to claim 1 when being executed by a processor.

5. An apparatus for real-time learning of a travelling strategy of a driverless vehicle, comprising:

one or more processors;

a storage device configured to store one or more programs;

wherein the one or more programs are executed by the one or more processors to enable the one or more processors to:

determine whether a first travelling strategy of the driverless vehicle is unable to be generated according to routing information depending on a high-precision map and surrounding obstacle information of the driverless vehicle;

in response to determining that the first travelling strategy of the driverless vehicle is unable to be generated according to the routing information depending on the high-precision map and the surrounding obstacle information, of the driverless vehicle, record travelling trajectories of other vehicles on a road;

when a number of vehicles on a same travelling trajectory is greater than a preset first number threshold, generate a second travelling strategy using the same travelling trajectory; and

control the driverless vehicle to travel using the second travelling strategy,

wherein the one or more programs are executed by the one or more processors to enable the one or more processors further to:

when a third travelling strategy is received by a second driverless vehicle from a server, determine whether the first travelling strategy is able to be generated by the second driverless vehicle;

in response to the first travelling strategy being unable to be generated by the second driverless vehicle, control the second driverless vehicle to travel using the third travelling strategy and transmit a used report indicating that the third travelling strategy has been used to the server, such that the server records a number of used reports and a duration of receiving used reports, and updates the high-precision map in response to the number of used reports is greater than a report threshold or the duration is longer than a duration threshold.

6. The apparatus according to claim 5 , wherein the one or more programs are executed by the one or more processors to enable the one or more processors to:

report the second travelling strategy to the server, and/or to transmit the second travelling strategy to at least one other driverless vehicle.

7. The apparatus according to claim 5 , wherein the one or more programs are executed by the one or more processors to enable the one or more processors to:

control the second driverless vehicle to travel using the first travelling strategy and transmit to the server a report indicating that the second driverless vehicle does not use the third travelling strategy, if the first travelling strategy is able to be generated by the second driverless vehicle.

8. A server for real-time learning of a travelling strategy of a driverless vehicle, comprising:

one or more processors;

a storage device configured to store one or more programs;

wherein the one or more programs are executed by the one or more processors to enable the one or more processors to:

receive a second travelling strategy from a first driverless vehicle, and obtain a position corresponding to the second travelling strategy; wherein the second travelling strategy is generated by: determining whether a first travelling strategy of the first driverless vehicle is unable to be generated according to routing information depending on a high-precision map and surrounding obstacle information, of the first driverless vehicle; in response to determining that the first travelling strategy of the first driverless vehicle is unable to be generated according to the routing information depending on the high-precision map and the surrounding obstacle information, of the first driverless vehicle, recording travelling trajectories of other vehicles on a road; when a number of vehicles on a same travelling trajectory is greater than a preset first number threshold, generating the second travelling strategy using the same travelling trajectory; and

when a number of received same second travelling strategies is greater than a preset second number threshold, determine the second travelling strategy as a third travelling strategy, and store the third travelling strategy and the corresponding position,

wherein the one or more programs are executed by the one or more processors to enable the one or more processors further to:

send the third travelling strategy to a second driverless vehicle, such that the second driverless vehicle determines whether the first travelling strategy is able to be generated, and is controlled to travel using the third travelling strategy in response to determining that the first travelling strategy being unable to be generated; and

in response to receiving used reports indicating that the third travelling strategy has been used, record a number of the used reports and a duration of receiving the used reports, and update the high-precision map in response to the number of used reports is greater than a report threshold or the duration is longer than a duration threshold.

9. The server according to claim 8 , wherein the one or more programs are executed by the one or more processors to enable the one or more processors to:

when a distance from a second driverless vehicle to a position corresponding to the third travelling strategy is detected to be less than a preset distance threshold, transmit the third travelling strategy to the second driverless vehicle.

10. The server according to claim 9 , wherein the one or more programs are executed by the one or more processors to enable the one or more processors to:

when a number of received reports from the second driverless vehicle indicating that the second driverless vehicle does not use the third travelling strategy exceeds a preset third number threshold, delete the third travelling strategy and the corresponding position.

11. The server according to claim 8 , wherein the one or more programs are executed by the one or more processors to enable the one or more processors to:

when a number of received reports from a second driverless vehicle indicating that the second driverless vehicle does not use the third travelling strategy exceeds a preset third number threshold, delete the third travelling strategy and the corresponding position.