Assistance method and assistance system and assistance device using assistance method that execute processing relating to a behavior model

An assistance device that executes a first processing including learning for a neural network of a first behavior model used for self-driving of a vehicle, the assistance device comprising:

• a processor;
• and
• a memory coupled to the processor, the memory storing a program that, when executed by the processor, causes the assistance device to:
• receive inputs of detected information from the vehicle, the detected information including information related to driver operations, the neural network of the first behavior model configured to be adjusted based on the information related to driver operations;
• determine two or more information items affecting accuracy of the first behavior model on the basis of the detected information, the two or more information items being selected from information items including a travel difficulty level of the vehicle, a wakefulness level of a driver, and a driving proficiency level of the driver;
• determine whether or not to execute the first processing including the learning for the neural network of the first behavior model based on levels of the two or more information items;
• execute the first processing including the learning for the neural network of the first behavior model in response to determining to execute the first processing, and not execute the first processing including the learning for the neural network of the first behavior model in response to determining to not execute the first processing;
• and
• output, to a reporting device on the vehicle, a determination result indicating whether or not the first processing including the learning for the neural network of the first behavior model is to be executed, and the levels of the two or more information items, the reporting device configured to concurrently display the determination result and the levels of the two or more information items.

An assistance method that executes a first processing including learning for a neural network of a first behavior model used for self-driving of a vehicle, the assistance method comprising:

• receiving inputs of detected information including information related to driver operations, the neural network of the first behavior model configured to be adjusted based on the information related to driver operations;
• determining, on the basis of the detected information thus input, two or more information items affecting accuracy of the first behavior model, the two or more information items being selected from information items including a travel difficulty level of the vehicle, a wakefulness level of a driver, and a driving proficiency level of the driver;
• determining, on the basis of levels of the two or more information items, whether or not to execute the first processing including the learning for the neural network of the first behavior model;
• executing the first processing including the learning for the neural network of the first behavior model in response to determining to execute the first processing, and not executing the first processing including the learning for the neural network of the first behavior model in response to determining to not execute the first processing;
• and
• outputting, to a reporting device on the vehicle, a determination result indicating whether or not the first processing including the learning for the neural network of the first behavior model is to be executed, and the levels of the two or more information items, the reporting device displaying the determination result and the levels of the two or more information items.