Scene Recognition And Path Planning Of Forklift AGV Equipment In Logistics Distribution Center Based On Visual Sensor

With the development of modern enterprises,all kinds of forklift AGV equipment are more and more widely used in distribution centers.The application of scene recognition and path planning in intelligent robots to forklift AGV equipment is of great significance to improve the automation and intelligence level of forklift AGV equipment.At present,the method of using camera to collect image information to obtain environmental information can break through the limitation of traditional guideway and positioning equipment,and is beneficial to the path planning and system expansion in the later stage of warehouse construction.Taking the forklift AGV equipment of the distribution center warehouse as the research object,this paper explores the scene identification and path planning of forklift AGV equipment based on deep convolution neural network,which has theoretical guiding significance for the intellectualization of distribution center warehousing system.First of all,through the analysis of the structure of the deep convolution neural network and the operation of each layer of the network,and based on the characteristics of the warehouse environment,the semantic segmentation network applied to the scene recognition in the warehouse environment is established.The semantic segmentation network combined with the characteristics of the depth separable convolution.The coding network of the semantic segmentation network is improved and optimized.The parameters of each layer of the semantic segmentation network are determined,and a scene recognition method suitable for warehouse environment is proposed.The equipment can use the deep learning method to learn the characteristics of the environment,achieve accurate recognition without adding environmental landmarks in a large-scale environment,and provide an effective convolution neural network model for the scene recognition of forklift AGV equipment in the warehouse environment.Secondly,the warehouse operating environment data set is established,and the data set with semantic label information is obtained by manual labeling based on the original data.Through the training of the model on the data set,the final network model is obtained.In the model training stage,the depth convolution neural network is converged by adjusting the training parameters.On this basis,the activation function layer of the model is studied.By using the activation function with better gradient performance,the nonlinear characteristics of the deep convolution neural network are increased,and the ability of the network model to express abstract features is improved At the same time,the improvement of activation function reduces the amount of calculation of the model in the training stage to a certain extent.Finally,a two-dimensional raster map is generated by using the label image obtained by semantic segmentation based on scene recognition of forklift AGV equipment.By analyzing the characteristics of warehouse equipment and working environment,the passable sector threshold strategy of improved VFH*algorithm is studied to realize the passability of equipment in warehouse environment,especially in aisle.Finally,the improved algorithm and the original algorithm are compared and analyzed.The applicability and effectiveness of the improved algorithm in the path planning of forklift AGV equipment in warehouse environment are verified.