Arithmetic operation device for vehicle

An automotive arithmetic device hosted on a motor vehicle, the automotive arithmetic device comprising:

• circuitry configured to
• estimate a vehicle external environment with a learned model that was trained by deep learning based on an output from an information acquisition device that acquires information of the vehicle external environment;
• calculate a first candidate route based on an estimation result previously obtained;
• recognize a target object outside the motor vehicle based on the output from the information acquisition device according to a predetermined rule set without using deep learning;
• set a static safe area which is an area where the target object is avoidable based on a recognition result that was previously provided by the circuitry;
• and
• determine a target motion of the motor vehicle by application of the first candidate route and the static safe area, wherein
• the circuitry
• sets the first candidate route as a route that the motor vehicle should take and determines the target motion of the motor vehicle so that the motor vehicle takes the first candidate route under a condition the first candidate route is entirely within the static safe area, and does not select the first candidate route as the route that the motor vehicle should take under a condition the first candidate route at least partially lies outside the static safe area,
• calculates a second candidate route that passes the static safe area only, and
• selects the second candidate route as the route that the motor vehicle should take and determines the target motion of the motor vehicle so the motor vehicle takes the second candidate route under a condition the first candidate route at least partially deviates from the static safe area.