Vehicle-to-vehicle ranging traveling method applied with reinforcement learning using a positioning sensor

A method for vehicle-to-vehicle ranging control applying reinforcement learning using a positioning sensor, the method comprising:

• determining one or more of location vectors, angles, or geometric triangles at a positioning node of each of a first vehicle that is turning and a second vehicle leading the first vehicle;
• determining a state vector of relative motion of the second vehicle with respect to the first vehicle;
• determining one or more of heading, a speed, or a yaw rate for the second vehicle;
• and
• performing, by a vehicle controller of the first vehicle, reinforcement learning for ranging correction between the first vehicle and the second vehicle as input data of deep learning, the reinforcement learning including using one or more of the state vector, the location vectors, the angles, or the geometric triangles that are determined based on the positioning node of each of the first vehicle and the second vehicle and one or more of the heading, the speed, or the yaw rate that are determined for the of the second vehicle as input data of deep learning.

A method for vehicle-to-vehicle ranging control applying reinforcement learning using a positioning sensor, the method comprising:

• determining one or more of location vectors, angles, or geometric triangles at a positioning node of each of a turning first vehicle and a second vehicle leading the turning first vehicle;
• determining a state vector based on relative motion of the second vehicle with respect to the turning first vehicle;
• determining one or more of heading, a speed, or a yaw rate for the second vehicle;
• and
• performing, by a server that communicates with the turning first vehicle and the second vehicle, reinforcement learning for ranging correction between the turning first vehicle and the second vehicle using one or more of the state vector, the location vectors, the angles, or the geometric triangles that are determined at the positioning nodes of the turning first vehicle and the second vehicle and one or more of the heading, the speed, or the yaw rate that are extracted from the second vehicle as input data of deep learning.