Road obstacle sensing method and system for vehicle multi-sensor fusion system

Abstract

The invention discloses a road obstacle sensing method and system for a vehicle multi-sensor fusion system. The method comprises the steps: acquiring or calculating four lane lines on the left side and the right side of a vehicle according to a set lane width; according to the four lane lines, dividing a self-vehicle lane where the vehicle is located and two bypass lanes; determining whether the bypass lanes are located in a drivable road area or not, if so, defining the area where the current bypass lanes are located as an area of interest, and if not, defining the area where the current bypass lanes are located as an observation area; when the multi-sensor fusion system works, detecting only the target objects with the risk of entering the area of interest in the observation region, anddetecting all the target objects in the area of interest. When the multi-sensor fusion system performs obstacle detection through the method, a large part of target objects which do not affect drivingpath planning can be filtered out, so that the load of the multi-sensor fusion system can be greatly reduced, and the working efficiency of the multi-sensor fusion system is improved.

Description

technical field [0001] The invention relates to the technical field of intelligent driving obstacle detection, in particular to a road obstacle sensing method and system for a vehicle multi-sensor fusion system. Background technique [0002] With the continuous development of communication and sensor technology, the intelligent driving technology of vehicles is gradually perfected. At this stage, the perception function of intelligent driving vehicles is mostly realized by multi-sensor fusion systems, and the sensors used in multi-sensor fusion systems are mostly cameras and radars, and these two sensors have a large amount of interference target data when detecting targets. The reasons are: [0003] 1. The radar can detect all obstacles on the road that the vehicle is driving that belong to the detection range of the sensor, and the sources of obstacles in the driving environment of the vehicle are complex and diverse, including vehicles, pedestrians, non-motorized vehicle...

AI Technical Summary

Problems solved by technology

[0004] 2. Due to the working characteristics of the radar itself, there may be false targets;

[0005] 3. During the driving process of the vehicle, the stability of the vehicle will be different under different road conditions, so the vehicle may be bumped or shaken. At this time, the detection signal of the radar may be lost, which will cause the target information to appear. large fluctuations

[0006] The input of these invalid target data and false target data will cause the ECU of the multi-sensor fusion system to increase the calculation load and reduce the processing efficiency.

Images