Font Size: a A A

Research On Path Navigation Of Intelligent Road Sweeper Based On Machine Vision

Posted on:2022-08-31Degree:MasterType:Thesis
Country:ChinaCandidate:J LiFull Text:PDF
GTID:2492306605461914Subject:Master of Engineering
Abstract/Summary:
In recent years,with the continuous improvement of China’s urbanization level,urban residents have become more and more,garbage and road surface area that need to be cleaned have gradually increased,and the work burden of sanitation vehicle drivers has become increasingly heavier.Although the sanitation industry is becoming more and more important,the traditional sanitation special vehicles have been difficult to meet the needs of urbanization.Driverless technology is on the rise,and machine vision technology used in navigation and path planning is also maturing.According to the cleaning requirements of low-speed road sweeper in a specific environment,this paper studies the road sweeper,and proposes a path navigation scheme of intelligent road sweeper based on machine vision,so that the road sweeper can drive autonomously-along the planned lane line.A QR code landmark is set at the garbage disposal point that needs to be parked temporarily.The camera detects and identifies the information in the QR code to realize the temporary parking of the vehicle.The active obstacle avoidance system is installed in the road sweeper to realize the emergency braking of the road sweeper to avoid accidents under special circumstances.Found in the actual testing process to extract the navigation line algorithm can well control the autonomous driving of intelligent sweep road car on the straight line but in bend due to reasons such as vehicle inertia is very difficult to realize smooth turning,and the obvious features of deep learning to bend the actual control effect is good,but the characteristic is not obvious linear part of the control effect is poorer.Therefore,different control algorithms are adopted in this paper to realize the control of the straight line part and the bend part.The test results show that the intelligent road sweeper can drive autonomously along the lane line and accurately recognize the information in the TWO-DIMENSIONAL code to complete the temporary parking task.The application scenario of this study is in industrial parks or factories with low human flow,which is significant to relieve the cleaning pressure,traffic pressure and personnel pressure brought by urbanization.
Keywords/Search Tags:Machine vision, path planning, qr code, road sweeper
Related items