Data writing method, recording medium, and vehicle
View Patent ↗A data writing method includes obtaining and writing. In the obtaining, a trained model constructed in advance by machine learning to estimate a state of a vehicle that is one of a human-powered vehicle or an electric bicycle is obtained. In the writing, the trained model obtained in the obtaining is written, in a nonvolatile memory area of a semiconductor memory, to a third area that is different from both a first area in which a bootloader is stored and a second area in which an application that references the trained model is stored.
1 . A data writing method comprising:
obtaining a trained model constructed in advance by machine learning to estimate a state of a vehicle that is one of a human-powered vehicle or an electric bicycle;
writing, in a nonvolatile memory area of a semiconductor memory, the trained model obtained in the obtaining to a third area that is different from both a first area in which a bootloader is stored and a second area in which an application that references the trained model is stored; and
controlling the vehicle by referencing the trained model stored in the third area of the semiconductor memory to execute the application stored in the second area,
wherein the semiconductor memory comprises a plurality of semiconductor memories, and
the plurality of semiconductor memories include a first semiconductor memory including the first area and the second area, and a second semiconductor memory including the third area.
2 . The data writing method according to claim 1 ,
wherein the trained model is a model that estimates an air pressure of a tire of the vehicle.
3 . The data writing method according to claim 1 ,
wherein in the obtaining, when an instruction to select a type of vehicle from among a plurality of types of vehicles is received during manufacture of the semiconductor memory, the trained model corresponding to the type of vehicle is obtained.
4 . The data writing method according to claim 1 ,
wherein in the obtaining, when an instruction to select a type of trained model from among a plurality of types of trained models is received after manufacture of the semiconductor memory, the type of trained model is obtained.
5 . A non-transitory computer-readable recording medium having recorded thereon a program for causing one or more processors to execute a data writing method comprising:
obtaining a trained model constructed in advance by machine learning to estimate a state of a vehicle that is one of a human-powered vehicle or an electric bicycle;
writing, in a nonvolatile memory area of a semiconductor memory, the trained model obtained in the obtaining to a third area that is different from both a first area in which a bootloader is stored and a second area in which an application that references the trained model is stored; and
controlling the vehicle by referencing the trained model stored in the third area of the semiconductor memory to execute the application stored in the second area,
wherein the semiconductor memory comprises a plurality of semiconductor memories, and
the plurality of semiconductor memories include a first semiconductor memory including the first area and the second area, and a second semiconductor memory including the third area.
6 . A vehicle that is one of a human-powered vehicle or an electric bicycle, the vehicle comprising:
a body;
a semiconductor memory to which data has been written by a trained model constructed in advance by machine learning to estimate a state of the vehicle, wherein the trained model is written, in a nonvolatile memory area of the semiconductor memory, to a third area that is different from both a first area in which a bootloader is stored and a second area in which an application that references the trained model is stored; and
a controller that controls the vehicle by referencing the trained model stored in the third area of the semiconductor memory to execute the application stored in the second area,
wherein the semiconductor memory comprises a plurality of semiconductor memories, and
the plurality of semiconductor memories include a first semiconductor memory including the first area and the second area, and a second semiconductor memory including the third area.