Map system, map generating program, storage medium, on-vehicle apparatus, and server

A non-transitory computer-readable storage medium storing a map generation program for causing at least either of a vehicle apparatus and a server to perform an integrating step, wherein

• the vehicle apparatus is mounted to each vehicle of a plurality of vehicles and includes an image sensor that captures an image of a surrounding area of a respective vehicle,
• the server generates a probe map using data that is transmitted from the respective vehicle, the data corresponding to feature points extracted from the image captured by the image sensor that is transmitted from the vehicle apparatus of the respective vehicle,
• the integrating step comprises:
• determining bias in the data in which pieces of the data are biased by data of limited accuracy from the image sensor of the respective vehicle, the limited accuracy of the image sensor being determined based on at least one of (i) at least one of a mounting position, a mounting attitude, and a resolution and an angle of view of the image sensor, (ii) a traveling speed of the vehicle, and (iii) a brightness of a surrounding environment of the vehicle;
• weighting pieces of data of a plurality of data that is transmitted from the plurality of vehicles based on the determined bias in the pieces of the data;
• and
• generating a map by integrating at least a portion of the pieces of data of the plurality of data that is transmitted from the plurality of vehicles based on the weighting,
• wherein
• the map is transmitted to at least one vehicle of the plurality of vehicles, and an actuator of the at least one vehicle performs a process for controlling the traveling of the at least one vehicle based on the map, wherein:
• for a state in which an estimation accuracy of an own-vehicle behavior is determined to be relatively high, the server weights the pieces of data such that a priority level of the pieces of data is higher than a priority level of the data in which the estimation accuracy is determined to be relatively low, and
• the estimation accuracy of the own-vehicle behavior being estimated using the respective image that corresponds to the pieces of data and estimated using a Structure From Motion method.

A vehicle apparatus that is mounted to each vehicle of a plurality of vehicles, the vehicle apparatus comprising:

• an image sensor that captures an image of a surrounding area of a respective vehicle of the plurality of vehicles, wherein:
• the vehicle apparatus transmits, to a server, data that corresponds to feature points extracted from the image captured by the image sensor, the server is configured to:
• determine bias in the data in which pieces of the data are biased by data of limited accuracy from the image sensor of the respective vehicle, the limited accuracy of the image sensor being determined based on at least one of (i) at least one of a mounting position, a mounting attitude, and a resolution and an angle of view of the image sensor, (ii) a traveling speed of the vehicle, and (iii) a brightness of a surrounding environment of the vehicle;
• weight pieces of data of a plurality of data that is transmitted from the plurality of vehicles based on the determined bias in the pieces of the data;
• and
• generate a map by integrating at least a portion of the pieces of data of the plurality of data that is transmitted from the plurality of vehicles based on the weighting,
• wherein
• the map is transmitted to at least one vehicle of the plurality of vehicles, and an actuator of the at least one vehicle performs a process for controlling the traveling of the at least one vehicle based on the map,
• the vehicle apparatus further comprises:
• a processor:
• a non-transitory computer-readable storage medium;
• and
• a set of computer-executable instructions stored on the non-transitory computer-readable storage medium that cause the processor to
• attach reliability level information that is information related to a reliability level of the data that corresponds to the image, and
• the set of computer-executable instructions further cause the processor to
• evaluate the reliability level based on an estimation accuracy when an own-vehicle behavior, which is a behavior of the vehicle, is estimated using the image that corresponds to the data and estimated using a Structure From Motion method, and
• generate the reliability level information based on evaluation results thereof.

A server that generates a map using data corresponding to feature points extracted from an image that is captured by an image sensor that is provided in a vehicle apparatus that is mounted to each vehicle of a plurality of vehicles, the image sensor being included in the vehicle apparatus of a respective vehicle and configured to capture an image of a surrounding area of the respective vehicle, the data being transmitted from the vehicle apparatus, the server being configured to:

• determine bias in the data in which pieces of the data are biased by data of limited accuracy from the image sensor of the respective vehicle, the limited accuracy of the image sensor being determined based on at least one of (i) at least one of a mounting position, a mounting attitude, and a resolution and an angle of view of the image sensor, (ii) a traveling speed of the vehicle, and (iii) a brightness of a surrounding environment of the vehicle;
• weight pieces of data of a plurality of data that is transmitted from the plurality of vehicles based on the determined bias in the pieces of the data;
• and
• generate a map by integrating at least a portion of the pieces of data of the plurality of data that is transmitted from the plurality of vehicles based on the weighting,
• wherein
• the map is transmitted to at least one vehicle of the plurality of vehicles, and an actuator of the at least one vehicle performs a process for controlling the traveling of the at least one vehicle based on the map, wherein:
• for a state in which an estimation accuracy of an own-vehicle behavior is determined to be relatively high, the server weights the pieces of data such that a priority level of the pieces of data is higher than a priority level of the data in which the estimation accuracy is determined to be relatively low, and
• the estimation accuracy of the own-vehicle behavior being estimated using the respective image that corresponds to the pieces of data and estimated using a Structure From Motion method.