Lane auxiliary method for controlling steering torque based on vehicle-mounted blind area visual scene analysis

Abstract

The invention relates to a lane auxiliary method for controlling steering torque based on vehicle-mounted blind area visual scene analysis. The method comprises the steps of (1) based on a depth convolution neural network, carrying out vehicle and road identification line conjoint analysis on vehicle blind area visual scene input, and outputting boundary information of a lane where a current vehicle is located and co-directional vehicle information of adjacent lanes; (2) estimating the relative distance between the boundary of the lane where a rear side vehicle and the vehicle are located andthe vehicle, and calculating the pre-collision time of the vehicle and the vehicle in a blind area and the pre-deviation time of the lane where the vehicle is located; (3) sending graded early warningto a driver through two interactive modes of sound and graphic interfaces; and (4) controlling the steering torque of the vehicle by a PID controller through an electronic steering system to enable the vehicle to return to the current driving lane when the vehicle has no lane change intention and is about to deviate from the driving lane or has lane change intention and has a collision risk. Theoccurrence of corresponding potential accidents can be actively avoided, and the active safety performance of automatic vehicle driving is improved.

Description

Technical field [0001] The present invention relates to the technical field of automobile automatic driving, and in particular to a lane assist method for controlling steering torque based on on-board blind spot visual scene analysis. Background technique [0002] Intelligence is one of the important trends in the development of the automotive industry today. It is expected that intelligent driving technology and systems will develop rapidly around the world from 2020 to 2030. The technical roadmap of most vehicle manufacturers in the field of intelligence is steadily transitioning and upgrading from L1 and L2 assisted driving systems to L3 and above assisted driving systems. For mass-produced models, the degree of functional integration and cost of assisted driving systems are the focus of vehicle manufacturers. Existing advanced assisted driving systems mostly rely on forward and blind spot vision or millimeter wave radar system inputs to provide corresponding early warni...

AI Technical Summary

Problems solved by technology

To sum up, the above two existing systems need to be based on different sensory inputs. The lane-level assisted driving function of a single system is not fully applied and the system output is limited to L0-level early warning applications. When the driver cannot make correct warning signals It is difficult to avoid the occurrence of potential traffic accidents when responding

Images