CPC B60W 60/0018 (2020.02) [B60W 50/14 (2013.01); B60W 60/0053 (2020.02); B60W 60/0059 (2020.02); B60W 2554/4049 (2020.02)] | 19 Claims |
1. A method with a Safety of the Intended Functionality (SOTIF) scene collection and a self-update mechanism, which is applied to a vehicle, the method comprising:
a situation judging step comprising configuring a self-driving system to judge whether a sensing control dataset generated by a sensor and a controller belongs to one of an unexpected intervention dataset and an accident scene dataset to generate a situation judgment result;
a scene collecting step comprising configuring the self-driving system to collect the sensing control dataset to establish a scene database according to the situation judgment result;
a modifying step comprising configuring the self-driving system to modify an algorithm of the sensor and the controller according to the scene database;
a verifying step comprising configuring one of the self-driving system and a cloud platform to perform a parallel calculation on the sensor and the controller that are modified to generate a verification output command and to compare the verification output command with a driver intervention control command to generate a comparison result; and
an updating step comprising configuring the one of the self-driving system and the cloud platform to update the algorithm of the sensor and the controller according to the comparison result, thereby allowing an updated output command generated by the sensor and the controller that are updated to correspond to the driver intervention control command;
wherein as the situation judging step judges that the sensing control dataset belongs to the accident scene dataset, the sensing control dataset collected by the scene collecting step belongs to a SOTIF scene;
wherein the one of the self-driving system and the cloud platform controls displacement of the vehicle according to the updated output command and the driver intervention control command, and the accident scene dataset comprises:
a miss-operation dataset representing a dataset generated as the self-driving system operates in a situation that the self-driving system does not need to operate.
|