Intelligent scheduling industrial robot logistics warehousing system

Abstract

The invention relates to the field of industrial automation, and particularly provides an intelligent scheduling industrial robot logistics warehousing system. Based on an algorithm of reinforcement learning and real-time environment perception, the robot is endowed with the capability of dynamically planning the carrying path, the carrying efficiency is improved, and the collision risk is reduced. Based on an algorithm of space optimization and an intelligent storage strategy, storage space management is redefined, the space utilization rate is increased, and the storage retrieval process is optimized. The algorithms push the system to be efficient, accurate and intelligent from multiple dimensions, and a revolution is brought to the logistics storage industry.

Description

Technical Field

[0001] The present invention belongs to the field of industrial automation, and particularly relates to an intelligent scheduling industrial robot logistics warehousing system. Background Art

[0002] With the rapid development of e-commerce and modern logistics, the scale of the logistics warehousing industry has been continuously expanding, and the requirements for warehousing efficiency and management level have also become higher and higher. As a key device for realizing the automation of logistics warehousing, industrial robots play an important role in the links of goods handling, storage, sorting, etc. However, there are still many problems in the application of industrial robots in the field of logistics warehousing, which restricts the further development of the logistics warehousing industry.

[0003] Inaccurate goods identification and positioning: There are a wide variety of goods in logistics warehousing, with different shapes, sizes, and packages, and the placement positions are disorderly. The traditional industrial robot goods identification and positioning system mainly relies on a single vision recognition technology, which is easily affected by factors such as lighting conditions, goods occlusion, and packaging material reflection, resulting in low recognition and positioning accuracy. In actual operation, situations often occur where the robot grabs the wrong goods or cannot accurately reach the goods position, which not only reduces work efficiency but also may cause goods damage and warehouse order chaos.

[0004] Unreasonable handling path planning: The logistics warehousing environment is complex, with various obstacles such as shelves, aisles, other equipment, and personnel activities. Most of the existing industrial robot handling path planning methods are based on pre-set maps and fixed algorithms, and cannot perceive environmental changes and the influence of dynamic obstacles in real time. When encountering emergencies, it is difficult for the robot to quickly adjust the handling path, resulting in low transportation efficiency and even collision accidents.

[0005] Insufficient utilization of warehousing space: In logistics warehousing, reasonable utilization of warehousing space can improve the storage capacity and operation efficiency of the warehouse. At present, during the process of goods storage by industrial robots, there is often a lack of effective planning and management of warehousing space, and it is unable to perform intelligent storage layout according to factors such as the size, weight, and storage requirements of goods. This often leads to problems such as waste of storage space and unreasonable stacking of goods in the warehouse, restricting the storage capacity and operation benefits of the warehouse. Summary of the Invention

[0006] The present invention provides an intelligent scheduling industrial robot logistics warehousing system, aiming to solve problems such as goods identification and positioning, handling path planning, warehousing space utilization, and goods sorting efficiency faced by industrial robots during the logistics warehousing process. By introducing advanced sensing technologies, intelligent algorithms, and optimized control strategies, the operation efficiency and management level of industrial robots in the field of logistics warehousing are improved to meet the requirements of the modern logistics warehousing industry for high efficiency, precision, and intelligence, including:

[0007] The precise identification and positioning module of goods based on multi-source information fusion and deep learning obtains multi-source information of goods by integrating multi-type sensors such as high-resolution vision sensors, lidar, ultrasonic sensors, and radio frequency identification (RFID) tags, constructs a comprehensive information model of goods using multi-source information fusion technology, and combines deep learning algorithms to establish a mapping relationship model between goods features and identification and positioning results to achieve precise identification and positioning of goods;

[0008] The dynamic handling path planning module based on reinforcement learning and real-time environment perception uses lidar and vision sensors to collect obstacle information and dynamic changes in the logistics warehousing environment to construct a real-time environment map, transforms the handling path planning problem into an intelligent decision-making process, and defines the state space, action space, and reward function, and uses reinforcement learning algorithms to enable the robot to learn the optimal dynamic handling path planning strategy;

[0009] The efficient utilization module of warehousing space based on space optimization and intelligent storage strategies establishes a warehousing space optimization model by analyzing the warehouse space structure, goods size and weight, and storage requirements, uses intelligent storage strategies to automatically allocate the optimal storage location for goods, and monitors the storage status and space utilization in real time to dynamically adjust the storage layout;

[0010] The efficient production module for goods sorting based on multi-robot cooperation and task assignment optimization can reasonably allocate tasks to multiple industrial robots for collaborative completion according to the quantity and type of goods sorting tasks, realizes information sharing and collaborative operation among robots by establishing a communication and cooperation mechanism, and has an intelligent task assignment optimization function to adjust the task assignment plan in real time;

[0011] The intelligent monitoring and management platform for the logistics warehousing process interacts with the data of the industrial robot logistics warehousing system, collects and stores various data in the logistics warehousing process in real time, uses big data analysis and visualization technology for analysis and display, and has intelligent early warning and fault diagnosis functions.

[0012] Furthermore, in the precise identification and positioning module of goods based on multi-source information fusion and deep learning, the multi-source data fusion module uses a multi-source data fusion algorithm based on deep learning to fuse and process sensor data, and uses deep learning algorithms such as convolutional neural networks to extract detailed features of goods.

[0013] Furthermore, in the precise identification and positioning module of goods based on multi-source information fusion and deep learning, the deep learning model training module uses convolutional neural networks or recurrent neural networks to learn and train a large amount of goods data to establish a goods identification and positioning model.

[0014] Furthermore, in the dynamic handling path planning module based on reinforcement learning and real-time environment perception, the state space includes the position and attitude of the robot, surrounding environment information, and task objectives, the action space includes various motion actions of the robot, and the reward function is designed based on factors such as the robot safely and efficiently completing the handling task and avoiding obstacles.

[0015] Furthermore, in the dynamic handling path planning module based on reinforcement learning and real-time environment perception, a deep Q-network or proximal policy optimization algorithm is used to model and train the handling path planning process.

[0016] Furthermore, in the warehousing space efficient utilization module based on space optimization and intelligent storage strategy, mathematical optimization algorithms such as genetic algorithms and particle swarm optimization algorithms are used to establish a warehousing space optimization model.

[0017] Furthermore, the system architecture includes a perception layer, a data processing layer, a decision-making layer, an execution layer, and an application layer; the perception layer is equipped with various sensors to obtain cargo and environment information and the warehouse environment status and transmit it to the data processing layer; the data processing layer includes modules such as multi-source data fusion, deep learning model training, real-time environment map construction, warehousing space optimization, and multi-robot cooperation control; the decision-making layer includes modules such as accurate cargo identification and positioning, dynamic handling path planning, efficient utilization of warehousing space, efficient production of cargo sorting, and intelligent monitoring and management; the execution layer is performed by industrial robots and feedbacks information; the application layer is an intelligent monitoring and management platform.

[0018] Furthermore, various environmental monitoring devices in the perception layer include temperature and humidity sensors and smoke sensors.

[0019] Furthermore, the multi-source data fusion module in the data processing layer fuses the data collected by multi-source sensors to construct a comprehensive cargo and environment information model, the deep learning model training module uses deep learning algorithms to learn and train cargo data to establish a cargo identification and positioning model, the real-time environment map construction module constructs a real-time environment map based on lidar and visual sensor data, the warehousing space optimization module establishes a warehousing space optimization model according to the warehouse space structure and cargo characteristics and formulates an intelligent storage strategy, and the multi-robot cooperation control module manages and controls the communication and cooperation between multiple robots.

[0020] Furthermore, in the decision-making layer, the precise cargo identification and positioning module uses the precise cargo identification and positioning algorithm to achieve precise cargo identification and positioning based on the results of multi-source data fusion and deep learning model training. The dynamic handling path planning module generates the optimal dynamic handling path according to the real-time environment map construction results combined with the reinforcement learning algorithm. The efficient utilization module of the storage space realizes the efficient utilization of the storage space and intelligent storage management according to the results of the storage space optimization module. The efficient production module for cargo sorting reasonably distributes the cargo sorting tasks according to the results of the multi-robot cooperation control module and realizes the cooperation control of the robots. The intelligent monitoring and management module collects, stores, analyzes, and displays the logistics warehousing process data to achieve intelligent monitoring and management.

[0022] Beneficial effects

[0023] In terms of cargo identification and positioning, the algorithm based on multi-source information fusion and deep learning has completely changed the limitations of traditional single vision recognition technology. In the past, complex factors in the logistics warehousing environment, such as light changes, cargo occlusion, and the reflection of packaging materials, seriously affected the recognition effect. Now, this algorithm integrates multiple sensors, fuses multi-source information, and then, with the help of the powerful analysis ability of deep learning, constructs a precise comprehensive cargo information model. This enables cargo identification to no longer be limited to surface features and can deeply explore the internal attributes of the cargo, thus achieving precise identification and positioning of various types of cargo. In actual operation, it has greatly reduced a series of problems caused by incorrect identification, such as ineffective handling, cargo damage, and warehouse order chaos, providing a solid accuracy foundation for warehousing operations and greatly improving the overall operation efficiency.

[0024] The dynamic handling path planning algorithm based on reinforcement learning and real-time environment perception has brought a qualitative leap to the handling link in logistics warehousing. Traditional fixed path planning algorithms are unable to cope with the dynamically changing warehousing environment. The new algorithm uses real-time environment perception technology, which is like endowing industrial robots with the ability of "real-time insight" and can quickly capture any changes in the environment. By continuously optimizing the handling strategy through reinforcement learning, the robot seems to have the wisdom of autonomous decision-making and can dynamically adjust the handling path according to the real-time environment. This not only greatly improves the handling efficiency but also significantly reduces the probability of collision accidents, ensuring operation safety and enabling the logistics transportation to operate efficiently and stably in a complex environment.

[0025] This efficient warehouse space utilization algorithm, based on space optimization and intelligent storage strategies, redefines warehouse space management. Traditional warehousing methods often overlook the matching of cargo characteristics with spatial structure, resulting in significant space waste. This algorithm deeply analyzes the warehouse's spatial structure and the various attributes of the cargo, and utilizes intelligent storage strategies to meticulously plan a dedicated "home" for each item. Real-time adjustments to the storage layout ensure that the warehouse maintains efficient storage. This not only improves storage space utilization but also optimizes the storage and retrieval process, making warehouse operations more organized and significantly enhancing operational efficiency.

[0026] In summary, these innovative algorithms have injected powerful impetus into the industrial robot logistics and warehousing system from multiple key dimensions, such as the accuracy of cargo identification and positioning, the flexibility of transportation path planning, and the efficiency of storage space utilization, pushing it to stride towards efficiency, precision, and intelligence, bringing new vitality and changes to the development of the logistics and warehousing industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 Flowchart of system operation principle. DETAILED DESCRIPTION

[0028] Example 1

[0029] A large food processing company owns a modern logistics and warehousing center covering an area of 30,000 square meters and equipped with 8,000 storage spaces. The center handles a large number of raw materials and finished goods in and out of the warehouse every day, with an average daily order volume of 30,000 orders. The goods include various food raw materials, packaging materials, and finished foods packaged in different specifications. Before the introduction of the industrial robot logistics and warehousing system of the present invention, the company faced problems such as inaccurate cargo identification, confusing transportation routes, serious waste of storage space, and low cargo sorting efficiency. These problems not only led to high operating costs, but also affected the continuity of production and the timely delivery of customer orders.

[0030] 2. System Implementation Process

[0031] (1) Implementation of cargo accurate identification and positioning module based on multi-source information fusion and deep learning

[0032] Sensor Deployment: A high-resolution color vision sensor with a pixel count of 5000×4000 is installed on the industrial robot, which can clearly capture the details of the surface of the goods. It is paired with a lidar, with a scanning frequency set at 15Hz and a ranging accuracy of up to ±3mm, used to obtain the contour and spatial position information of the goods. At the same time, an ultrasonic sensor with a detection range of 0.05 - 8m and an RFID reader capable of quickly reading tag information are equipped. Each piece of goods in the warehouse is pasted with an RFID tag with a unique code, covering key information such as the name, specification, and batch of the goods.

[0033] Data Fusion and Feature Extraction: The multi-source data fusion module uses a fusion network based on the attention mechanism to efficiently fuse the data collected by different sensors. For example, the image data obtained by the vision sensor, the point cloud data of the lidar, and the distance data of the ultrasonic sensor can accurately extract detailed features such as the type, shape, size, position, and identification of the goods through this fusion network. When processing a batch of raw materials for canned fruits, the brand, specification, and precise position of the cans on the shelf can be accurately identified by fusing the data.

[0034] Model Training and Application: The deep learning model training module has collected data on more than 800,000 different types of goods, constructing a rich training dataset. Using a convolutional neural network (CNN) to perform deep learning and training on this data, an accurate goods recognition and positioning model has been established. In actual operation, when the robot is performing the task of handling goods, by analyzing the multi-source information collected in real time through this model, the recognition accuracy has been improved from 65% of traditional single vision recognition to over 97%. For example, when identifying chocolate products in various packages, it can accurately determine the flavor, specification, and batch of the products.

[0035] (2) Implementation of the Dynamic Handling Path Planning Module Based on Reinforcement Learning and Real-Time Environment Perception

[0036] Real-Time Environment Map Construction: The lidar and vision sensors are used to collect the obstacle information and dynamic changes in the logistics and warehousing environment in real time. Through advanced point cloud processing and image processing technologies, a real-time environment map with an accuracy of 8mm is constructed. This map can reflect the shelf layout, aisle conditions, and dynamic positions of personnel and other equipment in the warehouse in real time and accurately. For example, when a warehouse worker temporarily places a box of goods in the aisle, the map can quickly update this obstacle information.

[0037] Reinforcement learning model training: Convert the handling path planning problem into an intelligent decision-making process, clearly define the state space, including the precise position of the robot (accuracy up to ±8mm), attitude (angle accuracy ±0.3°), surrounding environment information (obstacle distance accuracy ±15mm), and task objectives, etc.; the action space covers various motion actions of the robot such as forward, backward, turning, lifting, etc.; the reward function is carefully designed based on factors such as whether the robot can complete the handling task safely and efficiently and avoid obstacles. Use the Deep Q-Network (DQN) to model and train the handling path planning process, and the training duration is about 80 hours.

[0038] Dynamic path planning execution: During the actual handling process, the dynamic handling path planning module generates the optimal dynamic handling path based on the real-time environment map and the trained reinforcement learning model, combined with real-time environment information. For example, when a forklift operation suddenly appears on the originally planned path while the robot is handling flour raw materials, the system can re-plan the path within 0.5 seconds to ensure the smooth progress of the handling task. Compared with traditional fixed path planning, the handling efficiency has increased by 35%, and the average handling time has been shortened from the original 6 minutes per piece of goods to 3.9 minutes.

[0039] (III) Implementation of a highly efficient warehousing space utilization system based on space optimization and intelligent storage strategies

[0040] Warehouse and goods information collection: Through precise measurement and modeling, obtain detailed spatial structure information of the warehouse, such as the layout, dimensions (shelf height 12m, each layer height 0.6m), load-bearing capacity (each layer can bear 600kg), etc. At the same time, conduct a comprehensive classification and attribute registration of the goods in the warehouse, and record in detail information such as the size, weight, and storage requirements of the goods. For example, for perishable food raw materials, record their shelf life and storage temperature requirements; for large packaging materials, record their volume and shape.

[0041] Establishment of a warehousing space optimization model: Use the particle swarm optimization algorithm to establish a warehousing space optimization model. This model automatically assigns the optimal storage location for the goods according to their storage requirements. For example, for goods with frequent inbound and outbound, assign them to the shelf positions close to the aisle; for heavier goods, assign them to the bottom layer of the shelf. During the goods storage process, monitor the storage status and space utilization of the warehouse in real time, and dynamically adjust the storage layout. Through this system, the storage capacity of the warehouse has increased by 25%, from the original 6000 cubic meters to 7500 cubic meters, and the warehousing space utilization rate has increased from 55% to 70%.

[0042] (IV) Implementation of a highly efficient goods sorting production system based on multi-robot collaboration and task assignment optimization

[0043] Robot task allocation: According to the task volume and types of goods sorting, tasks are reasonably allocated to 15 industrial robots to complete collaboratively. For example, when processing an order containing various flavors of potato chips and beverages, the system will reasonably allocate different robots to be responsible for sorting different categories of goods based on the current positions and task execution capabilities of the robots. By establishing a communication and collaboration mechanism among multiple robots, information sharing and collaborative operations among robots are achieved.

[0044] Task allocation optimization: The system has an intelligent task allocation optimization function, which can adjust the task allocation plan in real time according to the changes in tasks and the working status of robots. When a certain robot fails or the task volume suddenly increases, the system can re-allocate tasks within 1 minute to ensure the efficient and stable operation of the sorting process. Compared with traditional manual sorting, the goods sorting efficiency has increased by 45%, and the average daily sorted order volume has increased from the original 25,000 orders to 36,250 orders.

[0045] (V) Implementation of the intelligent monitoring and management platform for the logistics warehousing process

[0046] Data collection and interaction: Build an intelligent monitoring and management platform, and through data interaction with the industrial robot logistics warehousing system, various data in the logistics warehousing process are collected and stored in real time, including goods information, handling paths, warehouse space utilization, robot operation status, etc. For example, the platform records in real time the operation trajectories, task completion progress, and health status of each robot.

[0047] Data analysis and early warning: Using big data analysis and visualization technologies, the collected data is deeply analyzed and intuitively displayed to provide real-time monitoring and decision-making support for management personnel in the logistics warehousing process. The platform also has intelligent early warning and fault diagnosis functions, which can timely detect abnormal situations in the logistics warehousing process, such as equipment failures, missing goods, etc., and give early warnings and diagnoses. Since the platform was put into use, the equipment fault response time has been shortened from the original average of 1.5 hours to within 10 minutes, effectively ensuring the smooth progress of the logistics warehousing process.

[0048] III. Summary of implementation effects

[0049] Goods identification and positioning: Through multi-source information fusion and deep learning algorithms, the identification accuracy has been significantly improved, jumping from 65% to over 97%, greatly reducing goods grasping errors and positioning deviations, improving handling accuracy, and reducing the risk of goods loss.

[0050] Handling path planning: Based on the dynamic handling path planning algorithm of reinforcement learning and real-time environment perception, the handling efficiency has been increased by 35% significantly, the average handling time has been effectively shortened, the logistics transportation efficiency has been significantly improved, and the time cost has been reduced.

[0051] Warehouse space utilization: By means of space optimization and intelligent storage strategies, the storage capacity of the warehouse has increased by 25%, the utilization rate of warehouse space has been significantly improved, the storage potential of the warehouse has been fully exploited, and the warehousing cost has been reduced.

[0052] Goods sorting: An efficient production system for goods sorting based on multi-robot collaboration and task assignment optimization. The goods sorting efficiency has increased by 45%, the daily average number of sorted orders has increased significantly, meeting the growing order processing needs of enterprises and enhancing customer satisfaction.

[0053] Overall operation: The establishment of an intelligent monitoring and management platform has greatly shortened the response time to equipment failures, effectively ensured the smooth progress of the logistics warehousing process, improved the stability and reliability of overall operations, and reduced operational risks.

[0054] Through the implementation and application in the logistics warehousing center of this food processing enterprise, the effectiveness and superiority of the industrial robot logistics warehousing system based on multi-source information fusion and intelligent scheduling of the present invention in actual scenarios have been fully verified, providing strong support for the efficient, precise, and intelligent development of the logistics warehousing industry.

Claims

1. An intelligent scheduling industrial robot logistics warehousing system, characterized in that, Including: A precise cargo identification and positioning module based on multi-source information fusion and deep learning. It acquires multi-source information of the cargo by integrating multiple types of sensors such as high-resolution visual sensors, lidar, ultrasonic sensors, and radio frequency identification (RFID) tags. It constructs a comprehensive information model of the cargo using multi-source information fusion technology and establishes a mapping relationship model between cargo features and identification and positioning results in combination with deep learning algorithms to achieve precise identification and positioning of the cargo; A dynamic handling path planning module based on reinforcement learning and real-time environment perception. It uses lidar and visual sensors to collect obstacle information and dynamic changes in the logistics warehousing environment to construct a real-time environment map, transforms the handling path planning problem into an intelligent decision-making process, and defines the state space, action space, and reward function. It enables the robot to learn the optimal dynamic handling path planning strategy using reinforcement learning algorithms; A highly efficient utilization module for warehousing space based on space optimization and intelligent storage strategies. It establishes a warehousing space optimization model by analyzing the warehouse space structure, cargo size, weight, and storage requirements, automatically assigns the optimal storage location for the cargo using intelligent storage strategies, and monitors the storage status and space utilization in real time to dynamically adjust the storage layout; A highly efficient production module for cargo sorting based on multi-robot collaboration and task assignment optimization. It can reasonably allocate tasks to multiple industrial robots according to the cargo sorting task volume and types for collaborative completion, realizes information sharing and collaborative operation among robots by establishing a communication and collaboration mechanism, and has an intelligent task assignment optimization function to adjust the task assignment plan in real time; An intelligent monitoring and management platform for the logistics warehousing process. It interacts with the data of the industrial robot logistics warehousing system, collects and stores various data in the logistics warehousing process in real time, analyzes and displays them using big data analysis and visualization technologies, and has intelligent early warning and fault diagnosis functions.

2. The industrial robot logistics warehousing system based on multi-source information fusion and intelligent scheduling according to claim 1, characterized in that In the precise cargo identification and positioning module based on multi-source information fusion and deep learning, the multi-source data fusion module uses a multi-source data fusion algorithm based on deep learning to fuse and process sensor data, and extracts detailed cargo features using deep learning algorithms such as convolutional neural networks.

3. The industrial robot logistics warehousing system based on multi-source information fusion and intelligent scheduling according to claim 1, characterized in that, In the precise cargo identification and positioning module based on multi-source information fusion and deep learning, the deep learning model training module uses convolutional neural networks or recurrent neural networks to learn and train a large amount of cargo data to establish a cargo identification and positioning model.

4. The industrial robot logistics warehousing system based on multi-source information fusion and intelligent scheduling according to claim 1, characterized in that In the dynamic handling path planning module based on reinforcement learning and real-time environment perception, the state space includes the position, posture, surrounding environment information, and task objectives of the robot, the action space includes various movement actions of the robot, and the reward function is designed based on factors such as the robot's safe and efficient completion of the handling task and avoidance of obstacles.

5. The industrial robot logistics warehousing system based on multi-source information fusion and intelligent scheduling according to claim 1, wherein In the dynamic handling path planning module based on reinforcement learning and real-time environment perception, the deep Q network or proximal policy optimization algorithm is used to model and train the handling path planning process.

6. The industrial robot logistics warehousing system based on multi-source information fusion and intelligent scheduling according to claim 1, characterized in that, In the highly efficient utilization module for warehousing space based on space optimization and intelligent storage strategies, mathematical optimization algorithms such as genetic algorithms and particle swarm optimization algorithms are used to establish a warehousing space optimization model.

7. The industrial robot logistics warehousing system based on multi-source information fusion and intelligent scheduling according to claim 1, characterized in that, The system architecture includes a perception layer, a data processing layer, a decision-making layer, an execution layer, and an application layer. The perception layer is equipped with various sensors to obtain information about goods and the environment, as well as the warehouse environmental conditions, and transmits them to the data processing layer. The data processing layer includes modules such as multi-source data fusion, deep learning model training, real-time environmental map construction, warehouse space optimization, and multi-robot collaborative control. The decision-making layer includes modules such as precise goods identification and positioning, dynamic handling path planning, efficient utilization of warehouse space, efficient production of goods sorting, and intelligent monitoring and management. The execution layer is responsible for the industrial robots to execute tasks and provide feedback information. The application layer is an intelligent monitoring and management platform.

8. The industrial robot logistics warehousing system based on multi-source information fusion and intelligent scheduling according to claim 7, characterized in that, In the perception layer, various environmental monitoring devices include temperature and humidity sensors, and smoke sensors.

9. The industrial robot logistics warehousing system based on multi-source information fusion and intelligent scheduling according to claim 7, characterized in that, In the data processing layer, the multi-source data fusion module fuses the data collected by multi-source sensors to construct a comprehensive information model of goods and the environment. The deep learning model training module uses deep learning algorithms to learn and train the goods data to establish a goods identification and positioning model. The real-time environmental map construction module constructs a real-time environmental map based on lidar and visual sensor data. The warehouse space optimization module establishes a warehouse space optimization model according to the warehouse space structure and goods characteristics and formulates an intelligent storage strategy. The multi-robot collaborative control module manages and controls the communication and collaboration between multiple robots.

10. The industrial robot logistics warehousing system based on multi-source information fusion and intelligent scheduling according to claim 7, characterized in that, In the decision-making layer, the precise goods identification and positioning module uses the precise goods identification and positioning algorithm to achieve precise goods identification and positioning based on the results of multi-source data fusion and deep learning model training. The dynamic handling path planning module generates the optimal dynamic handling path according to the results of real-time environmental map construction in combination with reinforcement learning algorithms. The efficient utilization of warehouse space module realizes the efficient utilization of warehouse space and intelligent storage management according to the results of the warehouse space optimization module. The efficient production of goods sorting module reasonably distributes the goods sorting tasks according to the results of the multi-robot collaborative control module and realizes robot collaborative control. The intelligent monitoring and management module collects, stores, analyzes, and displays the data of the logistics warehousing process to achieve intelligent monitoring and management.

Images