Vehicle-mounted barrier detection method based on radar data and image data fusion and deep learning

Abstract

The invention discloses a target detection algorithm based on intelligent equipment sensor data fusion and deep learning. Through fused radar point cloud data and camera data, data feature types capable of being sensed by a detection model are enriched. By performing model training on data channel fusion with different configuration, optimal channel configuration is selected, detection accuracy isimproved, and meanwhile calculation power consumption is lowered. By performing a test on real data, the channel configuration suitable for a real condition is determined, and the goals of processingfusion data by use of a Yolo deep convolutional neural network model and performing target barrier detection on a road scene is achieved.

Description

technical field [0001] The invention relates to a deep learning obstacle detection method based on radar and image data fusion, and belongs to the technical fields of sensor fusion, artificial intelligence and automatic driving. Background technique [0002] Self-driving vehicles perceive the vehicle's own operating conditions, road environment, and surrounding obstacles through a variety of sensors assembled around the body; through algorithms such as target recognition, lane detection, traffic signal recognition, and feasible area detection, the environment is recognized. Generate driving control commands, which not only drive the modified steering, acceleration and braking actuators to complete the self-driving process, the whole process includes: "perception-cognition-decision-making" dynamic environment interaction closed loop. [0003] Common Lidars for self-driving vehicles include iBeo and velodyne forward radars (4 and 8 lines) and 360-degree surround radars (16 lin...

AI Technical Summary

Problems solved by technology

However, optical camera imaging does not have distance information, and generally requires complex calculations to obtain inter-frame information based on algorithms, and the accuracy rapidly decays with the increase of the actual target distance

The lack of distance information limits the application of camera data to autonomous driving scenarios, and the relevant data information has only a low degree of confidence

At the same time, the optical camera is greatly affected by ambient light. When the weather, light and environment change, the imaging effect is unstable, and it is difficult to adapt to the high-speed movement and bumpy imaging requirements in the vehicle scene.

Images