The Design Of Crane Anti-collision System Based On ARM Embedded Vision

With the development of economic construction, the use of crane is increasing rapidly.This gives more convenience,as well as more damages, to the lifting in practice,especially the crane collision accident.As a result, how to make it more sufficient and safe to operate crane is becoming more and more important.This issue will focus on how to achieve this goal,which is to say,how to make it relax and easy for crane drivers.This issue will introduce the design of software and hardware in the anti-collision system, which is based on the ARM embedded vision. It also introduces the process of how to build the platform of hardware and software.As ARM embedded processor have many advantages,such as tiny,low cost of development,high stability and low energy cost, the system will take ARM11 development platform as the hardware basical platform,the Linux system is also transplant and programming.Then the article makes analysis on the characters and principle of infrared ranging,laser ranging,ultrasonic ranging,radar ranging, monocular vision and binocular vision measurement distance method of range,then we determin the ranging plan for the sake of the requirement of accuracy and real-time.Taking the crane industrial operating environment as application background this system utilizes the technology which is combined monocular vision with laser ranging to overcome the low liability and susceptibility to interference of single sensor,which ensures the stablization of the system.This anti-crush system could measure the distance between vehicle and the nearest obstacle automatically,and show them to the operator so that the operator could make judgement and take measures to avoid dangerous situation from happening.To summarise,this article concludes the advantages and disadvantages of system design,makes analysis of data measuring,concludes the reason of error,and points out the way of improving and completing the system.