| Robot technology has been widely used in warehousing,inspection,logistics,and other fields.The interconnection between robots is generally established in the local LAN,which makes it impossible to operate robots to complete tasks at a distance.Based on the Kube Edge platform and combining edge computing with robot perception and navigation technology,this paper proposes a robot environment awareness framework and a navigation framework based on Kube Edge to realize robot perception and navigation planning under cross-LAN conditions.The main work of this paper is as follows:1.The robot environment perception framework was built to carry out real-time monitoring of target information,temperature,and illumination during the robot’s moving process.Aiming at the problem of inaccurate target recognition caused by poor illumination and low image contrast in the night environment,a night target recognition algorithm and thermal imaging technology were used to deal with it.The operator can view the detection of the target in real time,as well as the temperature,lighting,and other information.Through experiments,it is proven that the deep learning algorithm and sensor function can be effectively used under the perception framework,and at the same time,the power consumption of the device can be reduced,and the robot can realize the joint perception of environmental information.2.The robot navigation framework is built to solve the multi-robot navigation problem under cross-LAN conditions and the cloud-edge collaborative navigation problem.It can simultaneously control multiple edge robots to perform navigation tasks in the cloud,and it can also unload different algorithms in the navigation package to the cloud to complete cloud-edge collaborative navigation tasks.The effectiveness of the navigation framework has been proven through many experiments.High concurrency and development potential.3.The proposed sensing framework and navigation framework are applied to the actual robot mobile platform,and the joint sensing experiment of robots and the navigation experiment of multiple robots are carried out,which proves the robustness and stability of the proposed framework in the actual environment. |
