A logistics robot control method and system based on artificial intelligence

Abstract

The present invention proposes a logistics robot control method and system based on artificial intelligence, including: the Internet of Things server obtains work task data, robot demand data, and status data of multiple logistics robots; based on each data, the first logistics robot is selected and the work task data is sent; the first logistics robot goes to the location of the transport object according to the task data, obtains its transportation data and surrounding environment data, and determines the first transportation plan considering the robot state; the first logistics robot sends the plan back to the Internet of Things server, and the Internet of Things server obtains sensor data collected by multiple edge intelligent sensors, modifies the first transportation plan into a second transportation plan according to these data, and sends it to the first logistics robot; the first logistics robot performs the transportation task of the transport object according to the second transportation plan. This solution uses environmental multi-source monitoring data to control logistics robots, improve operating efficiency, reduce costs, extend life, and enhance robust reliability.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an artificial intelligence-based logistics robot control method and system. Background Art

[0002] In practical applications, robots often need to move when performing tasks, such as logistics robots, guide robots, etc. There are currently two main methods for controlling robot movement. First, one is to control the movement of the robot through manual command, remotely control the robot through a remote control device, and avoid obstacles when they are detected; however, the labor cost of this method is high. Secondly, another method is to pre-set the robot's program, use the sensors installed on the robot to detect the surrounding environment, and stop or avoid obstacles when they are found. However, the detection accuracy of the sensor is limited, and it is impossible to fully understand the surrounding environment, which may cause the robot to frequently collide with obstacles, bringing the risk of causing serious damage to the robot. Summary of the invention

[0003] Based on the above problems, the present invention proposes a logistics robot control method and system based on artificial intelligence. Through the scheme of the present invention, the logistics robot can be controlled by utilizing multi-source monitoring data of the environment, which can improve the overall operating efficiency of the logistics robot, reduce operating costs, extend service life, improve system flexibility, enhance robust reliability, and realize intelligent automation.

[0004] In view of this, one aspect of the present invention proposes a logistics robot control method based on artificial intelligence, which is applied to a logistics robot control system based on artificial intelligence. The logistics robot control system based on artificial intelligence includes multiple logistics robots, multiple unmanned aircraft, multiple edge intelligent sensors, an Internet of Things server and a cloud server; wherein the logistics robot includes a central processing unit, a power management module, a positioning module and an obstacle removal module; the logistics robot control method based on artificial intelligence includes:

[0005] The IoT server obtains work task data and robot demand data corresponding to the work task data from the cloud server;

[0006] The Internet of Things server obtains robot status data of the plurality of logistics robots;

[0007] The Internet of Things server determines a corresponding first logistics robot from the plurality of logistics robots according to the robot status data, the work task data and the robot demand data, and selects a corresponding first work task data from the work task data and sends it to the first logistics robot;

[0008] The first logistics robot drives to the location of the first transport object according to the first work task data;

[0009] The first logistics robot obtains first transport object data of the first transport object and first surrounding environment data of the location of the first transport object, and determines a first transport plan according to the first transport object data, the first surrounding environment data, the first work task data and first robot state data of the first logistics robot;

[0010] The first logistics robot sends the first transportation plan to the Internet of Things server;

[0011] The Internet of Things server obtains first sensor data collected by the plurality of edge smart sensors;

[0012] The IoT server modifies the first transportation plan according to the first sensor data to obtain a second transportation plan, and sends the second transportation plan to the first logistics robot;

[0013] The first logistics robot transports the first transport object according to the second transport plan.

[0014] Optionally, the step of the IoT server acquiring work task data and robot demand data corresponding to the work task data from the cloud server includes:

[0015] The IoT server obtains corresponding work task data from the cloud server according to the unique identifier of the first logistics area managed by it;

[0016] The Internet of Things server sends the first area feature of the first logistics area to the cloud server;

[0017] The Internet of Things server receives the robot demand data determined and fed back by the cloud server according to the first area characteristics and the work task data.

[0018] Optionally, the IoT server determines a corresponding first logistics robot from the plurality of logistics robots according to the robot status data, the work task data and the robot demand data, and selects a corresponding first work task data from the work task data and sends it to the first logistics robot, comprising:

[0019] Preprocessing the robot state data to obtain sample robot state data;

[0020] Extracting task attribute data from the work task data;

[0021] identifying and quantifying said robot requirement data;

[0022] Based on the sample robot state data, a robot performance model is constructed using a machine learning algorithm to evaluate and quantify the operating capability of each logistics robot under different states;

[0023] Establishing a task requirement quantification model based on the task attribute data and the robot requirement data, mapping the task attribute data to actual requirement indicators, and scoring the task requirements based on cargo weight, time urgency, and environmental complexity;

[0024] Taking the robot performance model and the task requirement quantification model as input, designing a matching optimization objective function, and using a heuristic algorithm or a constrained optimization technique to solve an optimal allocation solution;

[0025] According to the optimal allocation plan, one or more of the logistics robots are selected as the first logistics robots to perform the task;

[0026] The corresponding first work task data is sent to the first logistics robot through the standard protocol.

[0027] Optionally, the first logistics robot acquires first transport object data of the first transport object and first surrounding environment data of the location of the first transport object, and determines a first transport plan according to the first transport object data, the first surrounding environment data, the first work task data and first robot state data of the first logistics robot, including:

[0028] The first logistics robot collects first transport object data of the first transport object through a sensor;

[0029] Collecting three-dimensional data of the surrounding environment of the location of the first transport object and performing three-dimensional reconstruction to obtain first surrounding environment data;

[0030] Extracting a first task constraint from the first work task data;

[0031] Acquire the first robot status data of the first logistics robot;

[0032] According to the first robot state data, based on a data-driven model, quantitatively evaluate the first transportation capacity of the first logistics robot in the current state;

[0033] The first transport object data, the first surrounding environment data, the first task constraint condition and the first transport capacity are input into a preset optimization algorithm, a multi-objective optimization function is constructed, and the optimal transport plan is solved using intelligent optimization technology to obtain a first transport plan.

[0034] Optionally, the step of the Internet of Things server acquiring the first sensor data collected by the plurality of edge intelligent sensors includes:

[0035] Determine a plurality of first influencing key points in the transportation process according to the first transportation plan;

[0036] Selecting a plurality of first edge intelligent sensors from the plurality of edge intelligent sensors according to the first influencing key point, the first transport object data, and the first robot state data, and determining a first data collection scheme for the first edge intelligent sensors;

[0037] Control the first edge intelligent sensor to collect data according to the first data collection scheme to obtain first sensor data.

[0038] Optionally, the step of determining a plurality of first influencing key points in the transportation process according to the first transportation plan includes:

[0039] Extracting a plurality of first key time nodes according to the time window requirements in the first transportation plan;

[0040] Based on the first transportation route planned in the first transportation plan, identifying a plurality of first geographical location points that meet preset geographical spatial characteristics;

[0041] Performing environmental modeling in combination with the three-dimensional spatial data and environmental data along the first transportation route, analyzing changes in environmental conditions along the route, and obtaining a first environmental condition change point;

[0042] Designing a pattern recognition algorithm based on rules or machine learning to automatically extract the first key influencing point that affects the transportation plan from the first key time node, the first geographical location point, and the first environmental condition change point;

[0043] In some possible implementations of the present invention, it also includes: for each of the first extracted influencing key points, determining a response strategy based on a preset robot model, associating the response strategy with the corresponding key point, and forming an influencing key point response plan.

[0044] Optionally, the step of selecting a plurality of first edge intelligent sensors from a plurality of edge intelligent sensors according to the first influencing key point, the first transport object data and the first robot state data, and determining a first data acquisition scheme for the first edge intelligent sensors includes:

[0045] Analyze a first monitoring target that needs to be monitored according to the type of the first influencing key point;

[0046] quantifying a first requirement for a sensor in combination with the first monitoring target, the first transport object characteristic in the first transport object data, and the data collection capability in the first robot state data;

[0047] Establishing a first sensor capability model for each of the edge intelligent sensors;

[0048] Matching the first requirement with a plurality of the first sensor capability models to find a sensor combination that meets the first requirement;

[0049] Design the optimization objective function and use the preset optimization method to solve the optimal sensor combination solution;

[0050] According to the optimization result, determining a plurality of first edge smart sensors to be used and the number thereof from the plurality of edge smart sensors;

[0051] Analyze the spatiotemporal distribution of the first key influencing points, and plan optimal layout positions for the plurality of the first edge smart sensors;

[0052] In combination with the state data of the first logistics robot, a working mode switching strategy of the first edge intelligent sensor during transportation is generated;

[0053] The selected plurality of the first intelligent sensors are networked, and data collection parameters, data upload strategies and destination ports are determined to obtain a first data collection solution.

[0054] Optionally, the step of the IoT server modifying the first transportation plan according to the first sensor data to obtain a second transportation plan, and sending the second transportation plan to the first logistics robot includes:

[0055] The IoT server configuration protocol parsing module performs data verification and reorganization on the first sensor data from each of the first edge smart sensors, and fuses and organizes the data according to time and space tags to form a complete key point area sensor data set;

[0056] Perform feature extraction and pattern recognition on the fused sensor data set to identify key environmental conditions that affect logistics and transportation;

[0057] Comparing the key environmental state with the first influencing key point in the first transportation plan to detect plan execution conflicts and risks caused by environmental changes;

[0058] Design an optimization algorithm for the conflicts and risks detected, and modify the first transportation plan under the constraints to form a new second transportation plan that meets the environmental changes.

[0059] Optionally, the step of the IoT server modifying the first transportation plan according to the first sensor data to obtain a second transportation plan, and sending the second transportation plan to the first logistics robot further includes:

[0060] The Internet of Things server determines a plurality of first unmanned aircraft from the plurality of unmanned aircraft according to the first sensor data;

[0061] The Internet of Things server organizes the plurality of the first unmanned aircraft into a first unmanned aircraft formation through a preset multi-aircraft formation control algorithm, so that the first unmanned aircraft formation flies along a preset path and interval to cover a target area, and monitors the target area;

[0062] Uploading the first three-dimensional environment data acquired by the first unmanned aircraft to the Internet of Things server through the airborne communication module;

[0063] The Internet of Things server integrates the first three-dimensional environmental data of all the first unmanned aircraft using a distributed perception fusion algorithm to reconstruct a first three-dimensional environmental model;

[0064] Planning an optimal route for the first unmanned aircraft formation to achieve full coverage monitoring of the target area according to the first three-dimensional environmental model;

[0065] The Internet of Things server receives the first unmanned aircraft perception data fed back by the first unmanned aircraft formation, modifies the first transportation plan according to the first unmanned aircraft perception data and the first sensor data to obtain a second transportation plan, and sends the second transportation plan to the first logistics robot.

[0066] Another aspect of the present invention provides a logistics robot control system based on artificial intelligence, comprising: a plurality of logistics robots, a plurality of unmanned aircraft, a plurality of edge intelligent sensors, an Internet of Things server and a cloud server; wherein the logistics robot comprises a central processing unit, a power management module, a positioning module and an obstacle removal module;

[0067] The IoT server obtains work task data and robot demand data corresponding to the work task data from the cloud server;

[0068] The Internet of Things server obtains robot status data of the plurality of logistics robots;

[0069] The Internet of Things server determines a corresponding first logistics robot from the plurality of logistics robots according to the robot status data, the work task data and the robot demand data, and selects a corresponding first work task data from the work task data and sends it to the first logistics robot;

[0070] The first logistics robot drives to the location of the first transport object according to the first work task data;

[0071] The first logistics robot obtains first transport object data of the first transport object and first surrounding environment data of the location of the first transport object, and determines a first transport plan according to the first transport object data, the first surrounding environment data, the first work task data and first robot state data of the first logistics robot;

[0072] The first logistics robot sends the first transportation plan to the Internet of Things server;

[0073] The Internet of Things server obtains first sensor data collected by the plurality of edge smart sensors;

[0074] The IoT server modifies the first transportation plan according to the first sensor data to obtain a second transportation plan, and sends the second transportation plan to the first logistics robot;

[0075] The first logistics robot transports the first transport object according to the second transport plan.

[0076] According to the technical solution of the present invention, the artificial intelligence-based logistics robot control method includes: an Internet of Things server obtains work task data and robot demand data corresponding to the work task data from a cloud server; the Internet of Things server obtains robot status data of multiple logistics robots; the Internet of Things server determines a corresponding first logistics robot from multiple logistics robots according to the robot status data, the work task data and the robot demand data, and selects the corresponding first work task data from the work task data and sends it to the first logistics robot; the first logistics robot travels to the location of the first transport object according to the first work task data; the first logistics robot obtains the first transport object data of the first transport object and the first surrounding environment data of the location of the first transport object, and determines the first transportation plan according to the first transport object data, the first surrounding environment data, the first work task data and the first robot status data of the first logistics robot; the first logistics robot sends the first transportation plan to the Internet of Things server; the Internet of Things server obtains the first sensor data collected by multiple edge intelligent sensors; the Internet of Things server modifies the first transportation plan according to the first sensor data to obtain a second transportation plan, and sends the second transportation plan to the first logistics robot; the first logistics robot carries the first transportation object according to the second transportation plan. Through the scheme of the present invention, the multi-source monitoring data of the environment can be used to control the logistics robot, which can improve the overall operating efficiency of the logistics robot (reasonable allocation can maximize the overall effectiveness of the robot fleet, avoid idle or excessive use of resources, and thus significantly improve the overall operating throughput and efficiency of the logistics system), reduce operating costs (optimized allocation schemes can minimize energy consumption, shorten operating time, etc., reduce the power consumption and maintenance costs of the robot, thereby reducing the operating costs of the logistics system), extend the service life (reasonable matching avoids excessive use of some robots, can balance the load intensity of each robot, and effectively extend the service life cycle of the entire robot fleet), improve system flexibility (AI optimization algorithms can quickly adapt to dynamic changes in the environment and demand, and adjust the allocation scheme in time, thereby improving the flexibility and responsiveness of the logistics system), enhance robust reliability (the optimization model quantifies the robot performance and task requirements, and the solved allocation scheme has a certain degree of fault tolerance and robustness, and can cope with abnormal situations), and realize intelligent automation (realizes the intelligent and automated scheduling of the logistics system, reduces manual intervention, and reduces operating labor costs). BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 is a flow chart of a logistics robot control method based on artificial intelligence provided by an embodiment of the present invention;

[0078] Figure 2 It is a schematic block diagram of an artificial intelligence-based logistics robot control system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0079] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0080] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0081] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0082] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0083] Refer to the following Figure 1 to Figure 2 To describe an artificial intelligence-based logistics robot control method and system provided according to some embodiments of the present invention.

[0084] like Figure 1 As shown, an embodiment of the present invention provides a logistics robot control method based on artificial intelligence, which is applied to a logistics robot control system based on artificial intelligence. The logistics robot control system based on artificial intelligence includes multiple logistics robots, multiple unmanned aircraft, multiple edge intelligent sensors, an Internet of Things server and a cloud server; wherein the logistics robot includes a central processing unit, a power management module, a positioning module and an obstacle removal module; the logistics robot control method based on artificial intelligence includes:

[0085] The IoT server obtains work task data and robot requirement data corresponding to the work task data (including but not limited to function, performance, endurance time, load capacity, safety, etc.) from the cloud server;

[0086] The IoT server obtains robot status data of the plurality of logistics robots (including but not limited to performance data, energy data, energy consumption data, safety data, load capacity data, driving stability data, etc.);

[0087] In this step, multiple sensors (such as current / voltage sensors, load sensors, inertial measurement units, etc.) can be integrated on each logistics robot; an embedded controller is used to collect various status data in real time, and pre-process the data such as filtering and denoising, and encapsulate it in a unified data format (such as JSON, XML or custom binary format); the logistics robot is equipped with a wireless communication module (such as WiFi, 4G / 5G, low-power wide area network, etc.), based on standard protocols such as MQTT and HTTP, and publishes status data to the IoT server regularly or according to event triggers, supporting disconnection reconnection, data caching and other mechanisms to ensure reliable data transmission; IoT server-side data Receiving and storing; the IoT server configures the message agent or Web service interface of the corresponding protocol, receives and parses the status data message from each robot, and stores the data in a distributed database or time series database according to dimensions such as timestamps; cleans and normalizes the massive status data; performs statistical analysis and modeling based on big data analysis technology (such as Spark); applies machine learning algorithms to detect and predict anomalies such as energy consumption and faults; builds a monitoring screen and report display interface based on Web visualization technology to intuitively present the overall status distribution and energy consumption trend of the logistics robot queue; provides decision support for scheduling decisions and maintenance decisions based on the analysis results. After implementing the above steps, the IoT server can efficiently obtain the distributed robot status big data and form a global perception of the robot queue, which has the following technical effects: improve the overall operational efficiency of the logistics system, optimize scheduling and maintenance strategies; enhance foresight and initiative, avoid faults, and reduce energy consumption; realize centralized monitoring, quickly locate and handle abnormal situations; data accumulation can support more value mining, such as health modeling, etc.; realize remote operation and maintenance, and reduce labor costs. In short, achieving all-round perception of the status of distributed logistics robots through industrial Internet of Things technology is the key to improving the level of intelligent, automated and visualized management of logistics systems.

[0088] The Internet of Things server determines a corresponding first logistics robot from the plurality of logistics robots according to the robot status data, the work task data and the robot demand data, and selects a corresponding first work task data from the work task data and sends it to the first logistics robot;

[0089] In this embodiment, the work task data includes but is not limited to: task ID (uniquely identifies a task), task type (such as handling, transportation, etc.), target location (task target item or the location that the robot needs to reach), target object / item (if it is a handling task, information about the item to be handled, such as name, weight, etc.), priority (task priority), completion time (estimated completion time of the task), starting location (the location where the robot obtains the task or the current location of the target item), route (if navigation is required, the route information specified by the task), status (the current status of the task, such as pending, in progress, completed, etc.), progress (if the task is carried out step by step, the current completion progress), log data (information that needs to be recorded during the task, such as location trajectory, etc.), etc.; through standardized data format and the necessary field information included, the details of the logistics robot task requirements can be effectively described and conveyed.

[0090] The first logistics robot drives to the location of the first transport object according to the first work task data;

[0091] In this step, the first logistics robot parses the target position, target object / item, completion time, starting position and other information from the first work task data; the logistics robot integrates a positioning module / system (such as GPS / BeiDou, laser SLAM, visual SLAM, etc.); obtains its own real-time and accurate position and posture in the global coordinate system or map, and obtains the target position coordinates of the transport object; based on the obtained robot and target positions, finds the optimal driving path from the map data (commonly used path planning algorithms include A*, D*, RRT*, SBPL and other variants), and the path planning needs to consider the starting point, end point constraints, obstacles, robot dynamics constraints, etc.; according to the planned path, derives the desired robot posture, speed and other motion trajectories; designs a motion controller (such as PID, model predictive control, etc.) to accurately track and adjust the robot motion; combines multi-sensor feedback, and corrects or re-plans in real time to cope with dynamic environmental changes; integrates sensor data such as vision and lidar to build a local environment map of the robot; detects and tracks static / dynamic obstacles in real time for intelligent obstacle avoidance; marks and locates the objects to be loaded and transported, and guides the last section of navigation. Through the above scheme, the robot can autonomously and orderly drive to the loading point, accurately align with the transport object, complete material loading, realize autonomous unmanned driving, and improve the efficiency and automation level of logistics operations; optimize the trajectory through AI planning algorithms to reduce the energy consumption of robot movement; intelligent perception and obstacle avoidance ensure driving safety and reduce the risk of collision; improve the accuracy of material loading through high-precision positioning and alignment; support dynamic path adjustment, have good environmental adaptability, and lay the foundation for further realization of collaborative operations and unmanned operations. In short, autonomous navigation is one of the core links in realizing intelligent logistics, and its accurate and reliable technical implementation plays a key role in improving logistics efficiency, reducing operating costs, and ensuring safety.

[0092] The first logistics robot obtains first transport object data of the first transport object and first surrounding environment data of the location of the first transport object, and determines a first transport plan according to the first transport object data, the first surrounding environment data, the first work task data and the first robot state data of the first logistics robot (the first transport plan includes but is not limited to the quantity, height, width, weight, center of gravity, time requirements, location, transportation route and three-dimensional spatial data and environmental data along the route, auxiliary equipment, safety protection plan, charging control plan, etc. of each transportation of goods);

[0093] The first logistics robot sends the first transportation plan to the Internet of Things server;

[0094] In this step, the logistics robot encodes the parameters related to the transportation plan in a predetermined format (such as JSON, XML or a custom binary format), and encapsulates the encoded data into a standard data message (optionally adding metadata such as timestamps, serial numbers, and error check codes, etc.); the logistics robot enables the wireless communication module (such as WiFi, 4G / 5G, low-power wide area network, etc.), and sends the data to the server through the wireless network based on standard protocols such as MQTT or private protocols; the transmission supports disconnection reconnection, data caching and other mechanisms to ensure reliable data arrival; the IoT server configures the message agent or transceiver program of the corresponding protocol, parses and verifies the legitimacy of the received data message, extracts the transportation plan parameters from the message, and performs persistent data storage; the IoT server integrates the transportation plan sent by the robot with other data sources, such as sorting out the plans of multiple robots for use by the upper-level scheduling system, and supports distribution to the superior management system or automatically executing corresponding operations according to needs. After completing the above steps, real-time two-way data interaction between the robot and the server can be realized. The IoT server can centrally manage and coordinate the transportation plans of multiple robots; support big data analysis to explore potential optimization space; facilitate remote monitoring and issue adjustment instructions based on global information; lay the foundation for subsequent automated scheduling, multi-robot collaboration, etc.; and improve the overall transparency and intelligence level of the logistics system.

[0095] The Internet of Things server obtains first sensor data collected by the plurality of edge smart sensors;

[0096] The IoT server modifies the first transportation plan according to the first sensor data to obtain a second transportation plan, and sends the second transportation plan to the first logistics robot;

[0097] The first logistics robot transports the first transport object according to the second transport plan.

[0098] The technical solution of this embodiment is adopted, and the method includes: the Internet of Things server obtains work task data and robot demand data corresponding to the work task data from a cloud server; the Internet of Things server obtains robot status data of multiple logistics robots; the Internet of Things server determines the corresponding first logistics robot from the multiple logistics robots according to the robot status data, the work task data and the robot demand data, and selects the corresponding first work task data from the work task data and sends it to the first logistics robot; the first logistics robot travels to the location of the first transport object according to the first work task data; the first logistics robot obtains the first transport object data of the first transport object and the first surrounding environment data of the location of the first transport object, and determines the first transportation plan according to the first transport object data, the first surrounding environment data, the first work task data and the first robot status data of the first logistics robot; the first logistics robot sends the first transportation plan to the Internet of Things server; the Internet of Things server obtains the first sensor data collected by multiple edge intelligent sensors; the Internet of Things server modifies the first transportation plan according to the first sensor data to obtain a second transportation plan, and sends the second transportation plan to the first logistics robot; the first logistics robot transports the first transportation object according to the second transportation plan. Through the scheme of the present invention, the multi-source monitoring data of the environment can be used to control the logistics robot, which can improve the overall operating efficiency of the logistics robot (reasonable allocation can maximize the overall effectiveness of the robot fleet, avoid idle or excessive use of resources, and thus significantly improve the overall operating throughput and efficiency of the logistics system), reduce operating costs (optimized allocation schemes can minimize energy consumption, shorten operating time, etc., reduce the power consumption and maintenance costs of the robot, thereby reducing the operating costs of the logistics system), extend the service life (reasonable matching avoids excessive use of some robots, can balance the load intensity of each robot, and effectively extend the service life cycle of the entire robot fleet), improve system flexibility (AI optimization algorithms can quickly adapt to dynamic changes in the environment and demand, and adjust the allocation scheme in time, thereby improving the flexibility and responsiveness of the logistics system), enhance robust reliability (the optimization model quantifies the robot performance and task requirements, and the solved allocation scheme has a certain degree of fault tolerance and robustness, and can cope with abnormal situations), and realize intelligent automation (realizes the intelligent and automated scheduling of the logistics system, reduces manual intervention, and reduces operating labor costs).

[0099] In some possible implementations of the present invention, the step of the IoT server acquiring work task data and robot demand data corresponding to the work task data from the cloud server includes:

[0100] The IoT server obtains corresponding work task data from the cloud server according to the unique identifier of the first logistics area managed by it;

[0101] The Internet of Things server sends the first area feature of the first logistics area to the cloud server;

[0102] The Internet of Things server receives the robot demand data determined and fed back by the cloud server according to the first area characteristics and the work task data.

[0103] In this embodiment, each IoT server manages a corresponding logistics area, and each logistics area is configured with a unique identifier; the cloud server determines and feeds back robot demand data (including but not limited to functions, performance, endurance time, load capacity, safety, etc.) based on the first area characteristics (including but not limited to spatial characteristics, humidity characteristics, temperature characteristics, road characteristics, etc.) corresponding to the first logistics area and the work task data. Through this solution, the robot demand data of the logistics robot that meets the logistics area characteristics and work task requirements can be accurately determined.

[0104] In some possible implementations of the present invention, the IoT server determines a corresponding first logistics robot from a plurality of logistics robots according to the robot status data, the work task data, and the robot demand data, and selects a corresponding first work task data from the work task data and sends it to the first logistics robot, comprising:

[0105] Preprocessing the robot state data (such as cleaning, normalization, etc.) to obtain sample robot state data;

[0106] Extracting task attribute data (such as operation location, time window, cargo characteristics, etc.) from the work task data;

[0107] Identify and quantify the robot's demand data (such as energy requirements, load requirements, safety requirements, functional requirements, etc.);

[0108] Based on the sample robot status data, a robot performance model is constructed using a machine learning algorithm to evaluate and quantify the operating capabilities (such as efficiency, stability, cruising range, etc.) of each logistics robot in different states;

[0109] Establishing a task requirement quantification model based on the task attribute data and the robot requirement data, mapping the task attribute data to actual requirement indicators, and scoring the task requirements based on cargo weight, time urgency, and environmental complexity;

[0110] In this step, the task attribute data refers to some basic characteristics of the task, such as the type, weight, size, etc. of the cargo, as well as the time limit (such as how long it must be completed), the working environment (such as in a narrow passage or in a spacious area), etc.; the robot demand data refers to the requirements for the robot's own functional capabilities, such as whether the battery life is enough to pull heavy objects, whether it can operate in a complex environment, etc.; combined with the above task attribute data and robot demand data, the task demand is quantitatively modeled, such as: constructing indicators including but not limited to "time urgency", "cargo weight and volume", "environmental complexity", "robot transportation capacity", etc.; then the task demand size can be judged by the cargo weight and environmental complexity, and then combined with the actual transportation capacity of the robot, a reasonable scoring evaluation is given to the specific demand. After the above process, by scoring, the corresponding solution demand size of different tasks can be quantified and compared, which will be conducive to task scheduling optimization and evaluation, and can reasonably quantify the size of the demand to avoid omissions or exaggeration; using standardized scoring indicators increases the rationality and comparability of the demand; can more comprehensively reflect the actual demand situation, and provide strong support for task decision-making. So to sum up, by quantitatively considering task attributes and machine requirements, we can more accurately and reasonably quantify the work task requirements and lay the foundation for future task decision-making and execution of intelligent systems.

[0111] Taking the robot performance model and the task requirement quantification model as input, designing a matching optimization objective function (such as minimizing energy consumption, maximizing work efficiency, etc.), and using a heuristic algorithm (such as a genetic algorithm) or a constrained optimization technique to solve the optimal allocation solution;

[0112] In this step, the robot performance model and task demand quantification model established previously are used as inputs of the optimization problem. The robot performance model describes the operating capability of each robot under different states; the task demand model quantifies the degree of demand for robots for each task; the optimization objective function is determined, which can be a single objective such as minimizing overall energy consumption, maximizing operating efficiency, etc., or a trade-off of multiple objectives, such as a weighted combination of energy consumption and efficiency; for this type of combined optimization problem, heuristic algorithms such as genetic algorithms, ant colony algorithms, simulated annealing, etc. can be used to obtain a better approximate solution; the robot, task model and optimization objective are input into the selected algorithm, and one or more optimal robot-task allocation schemes are solved under the constraints to make the objective function reach the extreme value. Through this process, robots and tasks can be optimally matched to maximize resource efficiency; the overall scheduling can be optimized to improve the overall operating efficiency of the logistics system; the optimization objectives can be flexibly set, such as focusing on energy saving or efficiency improvement; the AI ​​algorithm can handle large-scale task and robot matching combination problems; the allocation scheme has a certain robustness and can adapt to dynamically changing environments; and it lays the foundation for realizing autonomous and intelligent scheduling. In short, through mathematical modeling and optimization algorithms, we can solve a reasonable matching and allocation plan between robots and tasks, making the operation scheduling of the entire logistics system more intelligent and efficient, which is a key technical approach to improve the quality of logistics operations.

[0113] According to the optimal allocation plan, one or more of the logistics robots are selected as the first logistics robots to perform the task;

[0114] The corresponding first work task data is sent to the first logistics robot through the standard protocol.

[0115] In this embodiment, intelligent scheduling and operation planning based on artificial intelligence can be realized, and the following technical effects can be achieved: improving the overall efficiency of the logistics system, optimally matching robots and task requirements; dynamically balancing robot loads to avoid overuse or idleness of some robots; maximizing the service life of robots, reducing energy consumption and maintenance costs; improving the flexibility and adaptability of the logistics system to cope with dynamic changes; and laying the foundation for subsequent upgrades to centralized intelligent scheduling and multi-robot collaboration. Optimizing scheduling can not only improve the execution efficiency of a single task, but also ultimately improve the throughput and service quality of the overall system, which is a key technology for realizing intelligent logistics.

[0116] In some possible implementations of the present invention, the first logistics robot acquires first transport object data of the first transport object and first surrounding environment data of a location of the first transport object, and determines a first transport solution according to the first transport object data, the first surrounding environment data, the first work task data, and first robot state data of the first logistics robot, including:

[0117] The first logistics robot collects first transport object data (such as geometric size, weight, material, etc.) of the first transport object through a sensor;

[0118] Collecting three-dimensional data of the surrounding environment of the location of the first transport object and performing three-dimensional reconstruction to obtain first surrounding environment data (such as obstacles, spatial layout, etc.);

[0119] Extracting first task constraints (such as transportation requirements, time windows, path restrictions, etc.) from the first work task data;

[0120] Acquire the first robot status data of the first logistics robot (such as its own power, load, stability, data collection capability and other status data);

[0121] According to the state data of the first robot, based on the data-driven model, quantitatively evaluate the first transportation capacity of the first logistics robot in the current state (taking into account multiple influencing factors such as power consumption, obstacle avoidance, posture control, etc., to determine the transportation limit of the robot, such as maximum load, minimum turning radius, etc.);

[0122] The first transport object data, the first surrounding environment data, the first task constraint and the first transport capacity are input into a preset optimization algorithm, a multi-objective optimization function (such as maximizing efficiency, minimizing energy consumption, maximizing reliability, etc.) is constructed, and intelligent optimization technology (such as genetic algorithm, ant colony algorithm, etc.) is used to solve the optimal transportation plan to obtain a first transportation plan.

[0123] In this embodiment, the first transportation plan includes but is not limited to the quantity, height, width, weight, center of gravity, time requirements, location, transportation route and three-dimensional spatial data and environmental data along the route, auxiliary equipment, safety protection plan, charging control plan, etc. of each transportation of goods.

[0124] In this embodiment, it also includes: obtaining solutions such as the quantity of transported goods, the route, height restrictions, auxiliary facilities, etc. from the solved optimization results, combining safety protection strategies and power management to generate an executable transportation operation process; the first logistics robot starts loading and transporting materials according to the requirements of the plan, and monitors and evaluates the execution quality in real time.

[0125] In this embodiment, through the above steps, the robot can autonomously generate an efficient transportation plan that meets multiple constraints, maximize the robot's transportation capacity, and improve overall operating efficiency; reduce logistics energy consumption, extend the service life of batteries and robots; avoid potential safety hazards, and improve the reliability of the transportation process; cope with complex and changing environments, and demonstrate good adaptability and intelligence; lay the foundation for future multi-robot collaborative operations and dynamic optimization scheduling. In summary, intelligent transportation plan optimization is a key technology to improve the efficiency of unmanned logistics systems. Through the perception of comprehensive multi-source data and AI optimization decisions, the quality and efficiency of logistics transportation can be greatly improved.

[0126] In some possible implementations of the present invention, the step of the Internet of Things server acquiring the first sensor data collected by the plurality of edge intelligent sensors includes:

[0127] Determine multiple first influencing key points in the transportation process according to the first transportation plan (including but not limited to time points, special locations on the navigation route, areas where the goods are affected by the environment, etc.);

[0128] Selecting a plurality of first edge intelligent sensors from the plurality of edge intelligent sensors according to the first influencing key point, the first transport object data, and the first robot state data (including but not limited to determining the type, performance, location, quantity, etc. of the sensors), and determining a first data collection scheme for the first edge intelligent sensors;

[0129] Control the first edge intelligent sensor to collect data according to the first data collection scheme to obtain first sensor data.

[0130] In an embodiment of the present invention, a plurality of first influencing key points in the transportation process are determined by determining a first transportation plan, and corresponding edge intelligent sensors are deployed or enabled at the key points to accurately obtain monitoring data that requires focus during the transportation process.

[0131] In some possible implementations of the present invention, the step of determining a plurality of first influencing key points in the transportation process according to the first transportation plan includes:

[0132] According to the time window requirements in the first transportation plan, multiple first key time nodes are extracted (such as loading and unloading start / end time, transportation route congestion peak period, etc.; these time points may affect transportation efficiency, energy consumption, etc.);

[0133] Based on the first transport route planned in the first transport solution, identify multiple first geographical location points that meet preset geographical spatial characteristics (such as turning points, uphill and downhill sections, traffic intersections, narrow passages, etc. of the route; these locations may require the robot to slow down, perform posture adjustment, and other operations);

[0134] Perform environmental modeling in combination with the three-dimensional spatial data and environmental data along the first transportation route, analyze changes in environmental conditions along the route, and obtain first environmental condition change points (such as transition areas of environmental factors such as temperature and humidity, light, wind conditions, and ground conditions, where changes in the environment in these areas may affect cargo stability and robot working conditions);

[0135] Design a pattern recognition algorithm based on rules or machine learning to automatically extract the first influencing key points that affect the transportation plan from the first key time node, the first geographical location point, and the first environmental condition change point (the key point metadata includes parameters such as location, time, and impact type);

[0136] In some possible implementations of the present invention, it also includes: for each of the first influencing key points extracted, a response strategy (such as slowing down, changing routes, automatically turning on the dehumidification equipment, starting auxiliary sensors, etc.) is determined based on a preset robot model, and the response strategy is associated with the corresponding key point to form a response plan for the influencing key point.

[0137] In this embodiment, through the above steps, the logistics robot can identify all potential influencing factors in the transportation process in advance and generate targeted response plans, which can improve the adaptability and safety of the transportation process, reduce the probability of accidents, and reduce transportation risks; optimize energy consumption management, extend the overall transportation time and battery life; lay the foundation for highly automated transportation in complex environments, and improve the overall intelligence and autonomy of the logistics system. In short, the refined identification and response to key influencing points is an important part of realizing autonomous intelligent logistics, which can effectively improve the efficiency, reliability and environmental adaptability of transportation.

[0138] In some possible implementations of the present invention, the step of selecting a plurality of first edge intelligent sensors from a plurality of edge intelligent sensors according to the first influencing key point, the first transport object data, and the first robot state data, and determining a first data collection scheme for the first edge intelligent sensors includes:

[0139] According to the type of the first influencing key point (such as time point, path point, environmental area, etc.), analyzing the first monitoring target to be monitored (such as temperature and humidity, light, traffic flow, road conditions, etc.);

[0140] quantifying first requirements for sensors (such as requirements for accuracy and range) by combining the first monitoring target, first transport object characteristics (such as size, weight, sensitivity, etc.) in the first transport object data, and data acquisition capabilities (such as type and performance parameters of data acquisition modules, etc.) in the first robot state data;

[0141] Establishing a first sensor capability model of each edge intelligent sensor (including parameters such as measurement accuracy, detection range, working environment, power consumption, delay, etc.; modeling can be based on a physical model or a data-driven model);

[0142] Matching the first requirement with a plurality of the first sensor capability models to find a sensor combination that meets the first requirement;

[0143] Design optimization objective functions (such as minimizing cost, maximizing coverage, etc.), and use preset optimization methods (such as constrained optimization, integer programming, and other algorithms) to solve the optimal sensor combination solution;

[0144] According to the optimization result, determining a plurality of first edge smart sensors to be used and the number thereof from the plurality of edge smart sensors;

[0145] Analyze the spatiotemporal distribution of the first influencing key points, and plan optimal layout positions for the plurality of the first edge smart sensors;

[0146] In combination with the state data of the first logistics robot, a working mode switching strategy of the first edge intelligent sensor during transportation is generated;

[0147] The selected plurality of the first intelligent sensors are networked, and data acquisition parameters (such as acquisition time, sampling frequency, encoding format, etc.) are determined, as well as the data upload strategy and destination port, to obtain a first data acquisition solution.

[0148] In this embodiment, through the above steps, sensors are deployed on demand to improve resource utilization efficiency and reduce hardware investment; the key links of the entire process are covered to the maximum extent without omissions and blind spots; the sensor network is flexibly optimized to support intelligent scheduling and situational awareness; the overall observation quality meets the expected needs, reliably supports subsequent decisions, and lays a data foundation for intelligent analysis and timely intervention in the transportation process; it can promote the integrated innovation of sensor hardware and AI algorithms. In short, sensor network optimization and data collection strategy generation are important technical supports for realizing intelligent logistics, which are of great significance to ensuring transportation safety and quality and improving the level of logistics intelligence.

[0149] In some possible implementations of the present invention, the step of the IoT server modifying the first transportation plan according to the first sensor data to obtain a second transportation plan, and sending the second transportation plan to the first logistics robot includes:

[0150] The IoT server configuration protocol parsing module performs data verification and reorganization on the first sensor data from each of the first edge smart sensors, and fuses and organizes the data according to time and space tags to form a complete key point area sensor data set;

[0151] Perform feature extraction and pattern recognition on the fused sensor data set to identify key environmental conditions that affect logistics and transportation (such as traffic congestion, weather deterioration, etc.);

[0152] Compare the key environmental status with the first influencing key point in the first transportation plan to detect plan execution conflicts and risks caused by environmental changes (such as time conflicts caused by route congestion, weather changes affecting cargo safety, etc.);

[0153] In response to the detected conflicts and risks, an optimization algorithm (such as a genetic algorithm, etc.) is designed to modify the first transportation plan (such as adjusting and optimizing the transportation route, time schedule, etc.) while satisfying the constraints (such as the constraints of the first task) to form a new second transportation plan that meets the environmental changes.

[0154] In this embodiment, it also includes: sending the revised new transportation plan to the first logistics robot through the standard protocol; after receiving the new plan, the first logistics robot adjusts the transportation operation process in real time (such as rerouting, speed adjustment, etc.) to achieve closed-loop control of real-time perception of the edge environment and central decision-making feedback.

[0155] In this embodiment, through the above links, environmental changes can be discovered and responded to in a timely manner, unexpected risks can be avoided, and transportation reliability can be improved; transportation strategies can be dynamically optimized to maximize efficiency and reduce adverse effects; intelligent collaboration between the center and the edge can be achieved, the advantages of artificial intelligence can be brought into play, and the adaptability and robustness of the entire logistics system to the dynamic environment can be improved, laying the foundation for unmanned fully automatic transportation and scheduling in the future. In short, environmental perception and decision-making closed-loop control are the core capabilities of the IoT intelligent logistics system. By combining advanced AI algorithms with communication technologies, the level of intelligence and efficiency of logistics operations can be significantly improved.

[0156] In some possible implementations of the present invention, the IoT server modifies the first transportation plan according to the first sensor data to obtain a second transportation plan, and sends the second transportation plan to the first logistics robot, further comprising:

[0157] The Internet of Things server determines a plurality of first unmanned aircraft from the plurality of unmanned aircraft according to the first sensor data;

[0158] In this step, the first sensor data includes but is not limited to image data, temperature data, humidity data, sound data, light data, network connection data, etc. of the surrounding environment. Each unmanned aircraft can be equipped with different types of sensors, such as high-definition visual cameras, infrared thermal imagers, laser radars, multi-spectral imagers, etc., to obtain multi-source heterogeneous perception data such as visual, thermal imaging, and three-dimensional point clouds. The Internet of Things server can determine the assistance needs of the unmanned aircraft to assist in monitoring the environmental conditions in the current environment based on the first sensor data (including the target area for assisting monitoring and the specific tasks for assisting monitoring; such as determining the current spatial characteristics based on the image data, and further determining the blind area (target area) of the logistics robot and the edge intelligent terminal, so as to obtain the need for assisting in shooting blind area images; such as analyzing the coverage range of the corresponding edge intelligent sensor based on temperature data or humidity data or sound data or light data or network connection data, and assisting in monitoring the area (target area) outside the coverage range; the need for assisting in monitoring the key logistics area (target area) analyzed by the first sensor data), and multiple first unmanned aircraft can be determined from multiple unmanned aircraft according to the assistance needs.

[0159] The Internet of Things server organizes the plurality of the first unmanned aircraft into a first unmanned aircraft formation through a preset multi-aircraft formation control algorithm, so that the first unmanned aircraft formation flies along a preset path and interval to cover a target area, and monitors the target area;

[0160] Uploading the first three-dimensional environment data acquired by the first unmanned aircraft to the Internet of Things server through the airborne communication module;

[0161] The Internet of Things server integrates the first three-dimensional environmental data of all the first unmanned aircraft using a distributed perception fusion algorithm to reconstruct a first three-dimensional environmental model;

[0162] Planning an optimal route for the first unmanned aircraft formation to achieve full coverage monitoring of the target area according to the first three-dimensional environmental model;

[0163] The Internet of Things server receives the first unmanned aircraft perception data fed back by the first unmanned aircraft formation, modifies the first transportation plan according to the first unmanned aircraft perception data and the first sensor data to obtain a second transportation plan, and sends the second transportation plan to the first logistics robot.

[0164] In this embodiment, during route planning and task scheduling, the endurance of the unmanned aircraft itself is comprehensively considered, and the formation structure is dynamically adjusted to achieve efficient energy management.

[0165] In this embodiment, through the above technical route, the networked collaboration of unmanned aircraft can provide super-large robots with all-round, real-time, and high-precision environmental perception capabilities, realize three-dimensional detection and modeling of complex environments, effectively avoid obstacles, plan precise operation paths, greatly improve the intelligence and safety performance of robot operations, and avoid collision and failure risks. At the same time, it can also be used to monitor the robot's own status and perform remote auxiliary operations, which is a very promising collaborative paradigm.

[0166] In some possible implementations of the present invention, the IoT server organizes the plurality of the first unmanned aircraft into a first unmanned aircraft formation through a preset multi-aircraft formation control algorithm, so that the first unmanned aircraft formation flies along a preset path and interval to cover a target area, and the step of monitoring the target area includes:

[0167] Adopt distributed multi-robot formation control algorithms (such as behavior-based virtual structures, virtual leaders-followers, etc.) to establish a stable formation topology structure in the unmanned aircraft group and realize the coordinated flight control of the entire formation;

[0168] Design space filling algorithms (such as snake filling, grid partitioning, etc.) according to the geometry and operation requirements of the target area, automatically plan the optimal detection route of the formation, and maximize the coverage area and detection efficiency;

[0169] According to the detection capability of a single unmanned aircraft and the overall operational requirements of the formation, the detection tasks are reasonably divided, and the operation path and flight interval of each unmanned aircraft are determined to minimize duplication and omissions;

[0170] Through the field of view planning algorithm, each drone can adjust its viewing angle in time according to the detection target, and move in coordination with nearby drones to eliminate blind spots (improve detection accuracy and redundancy through overlapping fields of view);

[0171] Integrate distributed detection data and utilize the perception redundancy of unmanned aircraft formations to improve the detection and prediction capabilities of potential conflicts between dynamic obstacles and aircraft, and achieve collaborative obstacle avoidance and collision avoidance;

[0172] When a single unmanned aircraft fails, it can quickly detect and re-plan operations, dynamically adjust the formation structure, and achieve fault tolerance and self-recovery.

[0173] In this embodiment, through the above-mentioned series of collaborative control technologies, the unmanned aircraft formation can efficiently achieve full coverage detection of the target area. It not only has a large detection range and high timeliness, but also has field of view redundancy, collision avoidance safety and fault tolerance. It can obtain high-resolution three-dimensional environmental model data of key areas with high quality, provide accurate environmental perception support for ultra-large logistics robots, and greatly improve the intelligence and safety level of their operations.

[0174] It is understandable that in the application scenarios of logistics robots (especially large logistics robots), unmanned aircraft can play a very important complementary role, for example:

[0175] Unmanned aircraft can use onboard sensors (such as high-resolution cameras, lidar, etc.) to perform three-dimensional detection and real-time modeling of the operating environment, providing robots with high-precision environmental maps to guide planning and navigation;

[0176] From an aerial perspective, drones can monitor and supervise the operating status and operation progress of super-large robots in real time, detect abnormal situations in time, and ensure safe and efficient operations;

[0177] Unmanned aircraft can work together with super-large robots to perform tasks such as material handling, visual guidance, and obstacle removal, thus improving overall operational efficiency.

[0178] When the regional environment is harsh or communication is blocked, the drone can act as a relay node to maintain the transmission of data and control commands;

[0179] Once a logistics robot breaks down or gets stuck, the drone can quickly start up, approach the difficult area, and conduct inspections and emergency troubleshooting;

[0180] For some high-risk environments such as nuclear facilities and fire scenes, drones can replace logistics robots for preliminary surveys or disposal to avoid direct exposure risks;

[0181] Multiple unmanned aircraft can form a network and collaborate to act as the "eyes" and "ears" of super-large logistics robots and obtain comprehensive environmental information.

[0182] In summary, unmanned aircraft can act as the "attendants" and "assistants" of super-large logistics robots, which is conducive to improving the robots' environmental perception and collaboration capabilities, coping with various complex and dangerous situations, and enhancing the flexibility, safety and intelligence level of the entire system. The two can achieve good complementary synergy.

[0183] See also Figure 2 , Another embodiment of the present invention provides a logistics robot control system based on artificial intelligence, comprising: a plurality of logistics robots, a plurality of unmanned aircraft, a plurality of edge intelligent sensors, an Internet of Things server and a cloud server; wherein the logistics robot comprises a central processing unit, a power management module, a positioning module and an obstacle removal module;

[0184] The IoT server obtains work task data and robot demand data corresponding to the work task data from the cloud server;

[0185] The Internet of Things server obtains robot status data of the plurality of logistics robots;

[0186] The Internet of Things server determines a corresponding first logistics robot from the plurality of logistics robots according to the robot status data, the work task data and the robot demand data, and selects a corresponding first work task data from the work task data and sends it to the first logistics robot;

[0187] The first logistics robot drives to the location of the first transport object according to the first work task data;

[0188] The first logistics robot obtains first transport object data of the first transport object and first surrounding environment data of the location of the first transport object, and determines a first transport plan according to the first transport object data, the first surrounding environment data, the first work task data and first robot state data of the first logistics robot;

[0189] The first logistics robot sends the first transportation plan to the Internet of Things server;

[0190] The Internet of Things server obtains first sensor data collected by the plurality of edge smart sensors;

[0191] The IoT server modifies the first transportation plan according to the first sensor data to obtain a second transportation plan, and sends the second transportation plan to the first logistics robot;

[0192] The first logistics robot transports the first transport object according to the second transport plan.

[0193] It should be known that Figure 2 The block diagram of the artificial intelligence-based logistics robot control system is for illustration only, and the number of modules shown therein does not limit the protection scope of the present invention.

[0194] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0195] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0196] In the several embodiments provided in the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the above-mentioned units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0197] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0198] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0199] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a memory, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the above-mentioned methods of each embodiment of the present application. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or CD-ROM and other media that can store program codes.

[0200] A person skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable memory, and the memory can include: a flash drive, a read-only memory (English: Read-Only Memory, abbreviated as: ROM), a random access memory (English: Random Access Memory, abbreviated as: RAM), a magnetic disk or an optical disk, etc.

[0201] The embodiments of the present application are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of ​​the present application. At the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

[0202] Although the present invention is disclosed as above, the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions without departing from the spirit and scope of the present invention, and can make various changes and modifications, including the combination of the above-mentioned different functions and implementation steps, including software and hardware implementation methods, all of which are within the scope of protection of the present invention.

Claims

1. A logistics robot control method based on artificial intelligence, characterized in that: Applied to an artificial intelligence-based logistics robot control system, the artificial intelligence-based logistics robot control system includes multiple logistics robots, multiple unmanned aircraft, multiple edge intelligent sensors, an Internet of Things server and a cloud server; wherein the logistics robot includes a central processing unit, a power management module, a positioning module and an obstacle removal module; the artificial intelligence-based logistics robot control method includes: The IoT server obtains work task data and robot demand data corresponding to the work task data from the cloud server; The Internet of Things server obtains robot status data of the plurality of logistics robots; The Internet of Things server determines a corresponding first logistics robot from the plurality of logistics robots according to the robot status data, the work task data and the robot demand data, and selects a corresponding first work task data from the work task data and sends it to the first logistics robot; The first logistics robot drives to the location of the first transport object according to the first work task data; The first logistics robot obtains first transport object data of the first transport object and first surrounding environment data of the location of the first transport object, and determines a first transport plan according to the first transport object data, the first surrounding environment data, the first work task data and first robot state data of the first logistics robot; The first logistics robot sends the first transportation plan to the Internet of Things server; The Internet of Things server obtains first sensor data collected by the plurality of edge smart sensors; The IoT server modifies the first transportation plan according to the first sensor data to obtain a second transportation plan, and sends the second transportation plan to the first logistics robot; The first logistics robot carries the first transport object according to the second transport plan; The step of the Internet of Things server acquiring the first sensor data collected by the plurality of edge intelligent sensors includes: Determine a plurality of first influencing key points in the transportation process according to the first transportation plan; Selecting a plurality of first edge intelligent sensors from the plurality of edge intelligent sensors according to the first influencing key point, the first transport object data, and the first robot state data, and determining a first data collection scheme for the first edge intelligent sensors; Controlling the first edge intelligent sensor to collect data according to the first data collection scheme to obtain first sensor data; The step of determining a plurality of first influencing key points in the transportation process according to the first transportation plan includes: Extracting a plurality of first key time nodes according to the time window requirements in the first transportation plan; Based on the first transportation route planned in the first transportation plan, identifying a plurality of first geographical location points that meet preset geographical spatial characteristics; Performing environmental modeling in combination with the three-dimensional spatial data and environmental data along the first transportation route, analyzing changes in environmental conditions along the route, and obtaining a first environmental condition change point; Designing a pattern recognition algorithm based on rules or machine learning to automatically extract the first key influencing point that affects the transportation plan from the first key time node, the first geographical location point, and the first environmental condition change point; For each of the first extracted key points of influence, a response strategy is determined based on a preset robot model, and the response strategy is associated with the corresponding key point to form a response plan for the key points of influence; The step of selecting a plurality of first edge intelligent sensors from a plurality of edge intelligent sensors according to the first influencing key point, the first transport object data and the first robot state data, and determining a first data collection scheme for the first edge intelligent sensors includes: Analyze a first monitoring target that needs to be monitored according to the type of the first influencing key point; quantifying a first requirement for a sensor in combination with the first monitoring target, the first transport object characteristic in the first transport object data, and the data collection capability in the first robot state data; Establishing a first sensor capability model for each of the edge smart sensors; Matching the first requirement with a plurality of the first sensor capability models to find a sensor combination that meets the first requirement; Design the optimization objective function and use the preset optimization method to solve the optimal sensor combination solution; According to the optimization result, determining a plurality of first edge smart sensors to be used and the number thereof from the plurality of edge smart sensors; Analyze the spatiotemporal distribution of the first key influencing points, and plan optimal layout positions for the plurality of the first edge smart sensors; In combination with the state data of the first logistics robot, a working mode switching strategy of the first edge intelligent sensor during transportation is generated; The selected plurality of first edge intelligent sensors are networked, and data collection parameters, data upload strategies and destination ports are determined to obtain a first data collection solution.

2. The artificial intelligence-based logistics robot control method according to claim 1 is characterized in that: The step of the Internet of Things server acquiring work task data and robot demand data corresponding to the work task data from the cloud server includes: The IoT server obtains corresponding work task data from the cloud server according to the unique identifier of the first logistics area managed by it; The Internet of Things server sends the first area feature of the first logistics area to the cloud server; The Internet of Things server receives the robot demand data determined and fed back by the cloud server according to the first area characteristics and the work task data.

3. The artificial intelligence-based logistics robot control method according to claim 2 is characterized in that: The step of the IoT server determining a corresponding first logistics robot from the plurality of logistics robots according to the robot status data, the work task data and the robot demand data, and selecting a corresponding first work task data from the work task data and sending it to the first logistics robot comprises: Preprocessing the robot state data to obtain sample robot state data; Extracting task attribute data from the work task data; identifying and quantifying said robot requirement data; Based on the sample robot state data, a robot performance model is constructed using a machine learning algorithm to evaluate and quantify the operating capability of each logistics robot under different states; Establishing a task requirement quantification model based on the task attribute data and the robot requirement data, mapping the task attribute data to actual requirement indicators, and scoring the task requirements based on cargo weight, time urgency, and environmental complexity; Taking the robot performance model and the task requirement quantification model as input, designing a matching optimization objective function, and using a heuristic algorithm or a constrained optimization technique to solve an optimal allocation solution; According to the optimal allocation plan, one or more of the logistics robots are selected as the first logistics robots to perform the work task; The corresponding first work task data is sent to the first logistics robot through the standard protocol.

4. The artificial intelligence-based logistics robot control method according to claim 3 is characterized in that: The first logistics robot acquires first transport object data of the first transport object and first surrounding environment data of a location of the first transport object, and determines a first transport plan according to the first transport object data, the first surrounding environment data, the first work task data, and first robot state data of the first logistics robot, including: The first logistics robot collects first transport object data of the first transport object through a sensor; Collecting three-dimensional data of the surrounding environment of the location of the first transport object and performing three-dimensional reconstruction to obtain first surrounding environment data; Extracting a first task constraint from the first work task data; Acquire the first robot status data of the first logistics robot; According to the first robot state data, based on a data-driven model, quantitatively evaluate the first transportation capacity of the first logistics robot in the current state; The first transport object data, the first surrounding environment data, the first task constraint condition and the first transport capacity are input into a preset optimization algorithm, a multi-objective optimization function is constructed, and the optimal transport plan is solved using intelligent optimization technology to obtain a first transport plan.

5. The artificial intelligence-based logistics robot control method according to claim 4 is characterized in that: The step of the IoT server modifying the first transportation plan according to the first sensor data to obtain a second transportation plan, and sending the second transportation plan to the first logistics robot includes: The IoT server configuration protocol parsing module performs data verification and reorganization on the first sensor data from each of the first edge smart sensors, and fuses and organizes the data according to time and space tags to form a complete key point area sensor data set; Perform feature extraction and pattern recognition on the fused sensor data set to identify key environmental conditions that affect logistics and transportation; Comparing the key environmental state with the first influencing key point in the first transportation plan to detect plan execution conflicts and risks caused by environmental changes; Design an optimization algorithm for the conflicts and risks detected, and modify the first transportation plan under the constraints to form a new second transportation plan that meets the environmental changes.

6. The artificial intelligence-based logistics robot control method according to claim 5 is characterized in that: The step of the IoT server modifying the first transportation plan according to the first sensor data to obtain a second transportation plan, and sending the second transportation plan to the first logistics robot also includes: The Internet of Things server determines a plurality of first unmanned aircraft from the plurality of unmanned aircraft according to the first sensor data; The Internet of Things server organizes the plurality of the first unmanned aircraft into a first unmanned aircraft formation through a preset multi-aircraft formation control algorithm, so that the first unmanned aircraft formation flies along a preset path and interval to cover a target area, and monitors the target area; Uploading the first three-dimensional environment data acquired by the first unmanned aircraft to the Internet of Things server through the airborne communication module; The Internet of Things server integrates the first three-dimensional environmental data of all the first unmanned aircraft using a distributed perception fusion algorithm to reconstruct a first three-dimensional environmental model; Planning an optimal route for the first unmanned aircraft formation to achieve full coverage monitoring of the target area according to the first three-dimensional environmental model; The Internet of Things server receives the first unmanned aircraft perception data fed back by the first unmanned aircraft formation, modifies the first transportation plan according to the first unmanned aircraft perception data and the first sensor data to obtain a second transportation plan, and sends the second transportation plan to the first logistics robot.

7. A logistics robot control system based on artificial intelligence, characterized in that: include: Multiple logistics robots, multiple unmanned aircraft, multiple edge intelligent sensors, IoT servers and cloud servers; wherein the logistics robots include a central processing unit, a power management module, a positioning module and an obstacle removal module; The IoT server obtains work task data and robot demand data corresponding to the work task data from the cloud server; The Internet of Things server obtains robot status data of the plurality of logistics robots; The Internet of Things server determines a corresponding first logistics robot from the plurality of logistics robots according to the robot status data, the work task data and the robot demand data, and selects a corresponding first work task data from the work task data and sends it to the first logistics robot; The first logistics robot drives to the location of the first transport object according to the first work task data; The first logistics robot obtains first transport object data of the first transport object and first surrounding environment data of the location of the first transport object, and determines a first transport plan according to the first transport object data, the first surrounding environment data, the first work task data and first robot state data of the first logistics robot; The first logistics robot sends the first transportation plan to the Internet of Things server; The Internet of Things server obtains first sensor data collected by the plurality of edge smart sensors; The IoT server modifies the first transportation plan according to the first sensor data to obtain a second transportation plan, and sends the second transportation plan to the first logistics robot; The first logistics robot carries the first transport object according to the second transport plan; The step of the Internet of Things server acquiring the first sensor data collected by the plurality of edge intelligent sensors includes: Determine a plurality of first influencing key points in the transportation process according to the first transportation plan; Selecting a plurality of first edge intelligent sensors from the plurality of edge intelligent sensors according to the first influencing key point, the first transport object data, and the first robot state data, and determining a first data collection scheme for the first edge intelligent sensors; Controlling the first edge intelligent sensor to collect data according to the first data collection scheme to obtain first sensor data; The step of determining a plurality of first influencing key points in the transportation process according to the first transportation plan includes: Extracting a plurality of first key time nodes according to the time window requirements in the first transportation plan; Based on the first transportation route planned in the first transportation plan, identifying a plurality of first geographical location points that meet preset geographical spatial characteristics; Performing environmental modeling in combination with the three-dimensional spatial data and environmental data along the first transportation route, analyzing changes in environmental conditions along the route, and obtaining a first environmental condition change point; Designing a pattern recognition algorithm based on rules or machine learning to automatically extract the first key influencing point that affects the transportation plan from the first key time node, the first geographical location point, and the first environmental condition change point; For each of the first extracted key points of influence, a response strategy is determined based on a preset robot model, and the response strategy is associated with the corresponding key point to form a response plan for the key points of influence; The step of selecting a plurality of first edge intelligent sensors from a plurality of edge intelligent sensors according to the first influencing key point, the first transport object data and the first robot state data, and determining a first data collection scheme for the first edge intelligent sensors includes: Analyze a first monitoring target that needs to be monitored according to the type of the first influencing key point; quantifying a first requirement for a sensor in combination with the first monitoring target, the first transport object characteristic in the first transport object data, and the data collection capability in the first robot state data; Establishing a first sensor capability model for each of the edge smart sensors; Matching the first requirement with a plurality of the first sensor capability models to find a sensor combination that meets the first requirement; Design the optimization objective function and use the preset optimization method to solve the optimal sensor combination solution; According to the optimization result, determining a plurality of first edge smart sensors to be used and the number thereof from the plurality of edge smart sensors; Analyze the spatiotemporal distribution of the first key influencing points, and plan optimal layout positions for the plurality of the first edge smart sensors; In combination with the state data of the first logistics robot, a working mode switching strategy of the first edge intelligent sensor during transportation is generated; The selected plurality of first edge intelligent sensors are networked, and data collection parameters, data upload strategies and destination ports are determined to obtain a first data collection solution.

Images