Automatic navigation trolley system based on convolutional neural network and trolley route-following method

Abstract

The invention discloses an automatic navigation trolley system based on a convolutional neural network and a trolley route-following method. A camera of an AGV trolley acquires and transmits videos ofthe front of the AGV trolley to a control chip. A guiding mechanism comprises a rectangular net on the road surface, and road signs are attached to cross points of the rectangular net. A remote server transmits the structure of the rectangular net, road sign coordinate information and AGV trolley route information in a wireless communication mode. The AGV trolley acquires road sign information infront of the AGV trolley through the camera from a starting point and runs by comparing with the route information through the control chip. Road sign identification is performed through the camera on the AGV trolley without ultrasonic, laser infrared and other complicated devices and magnetic channel laying on the road surface, the device utilization rate is high, the function of identifying theroad signs with eyes can be simulated by utilizing the convolutional neural network to the greatest degree, the route-following precision can be constantly optimized and improved with optimization ofthe convolutional neural network, and the automatic navigation trolley system has very high research and development depth and market prospect.

Description

technical field [0001] The invention relates to an automatic navigation car system and a car line following method based on a convolutional neural network. Background technique [0002] AGV trolley refers to a transport vehicle equipped with automatic guidance devices such as electromagnetic or optical, capable of driving along a prescribed guiding path, with safety protection and various transfer functions. The whole process of AGV is Automated Guided Vehicle. The AGV trolley can be installed with different ends to complete the handling of workpieces of various shapes and states. It can be widely used in machine tool loading and unloading, stamping machine automatic production lines, automatic assembly lines, palletizing, and container automatic handling, which greatly reduces human labor. Heavy manual labor has broad market prospects. [0003] The development history of my country's moving vehicles is relatively late. In the 1950s, enterprise handling still mainly used o...

AI Technical Summary

Problems solved by technology

[0004] At present, automatic unmanned trucks controlled by computers have been launched on the market. The computer sends handling instructions to control the driving route of the car. The guiding methods of the car along the line mainly include electromagnetic induction guidance, ultrasonic guidance, laser guidance or infrared guidance. Considering Power consumption and cost issues, ultrasonic guidance, laser guidance, infrared guidance matching products are expensive, cost-effective, not easy to promote, most existing cars are guided by electromagnetic induction, and multiple magnetic tracks need to be laid on the road for navigation. However, the laying of multiple magnetic strips requires the reconstruction of the road surface, the space reuse rate is low, additional hardware and engineering costs are added, and the actual line tracking efficiency is affected.

Images