A logistics warehouse intelligent control method based on wireless signal positioning

Abstract

The application discloses a kind of logistics warehouse intelligent control methods based on wireless signal positioning, comprising the following steps: S1) signal receiving base station is installed on warehouse top, signal sending label is installed on forklift and goods pallet, and the signal receiving base station is connected with computer by network cable;S2) the signal sending label timing broadcast acceleration state to signal receiving base station, for positioning and behavior notification;S3) the computer receives from signal receiving base station's concurrent data queue, determines goods loading and unloading process and goods location.The logistics warehouse intelligent control method based on wireless signal positioning provided by the application can realize low-cost, rapid deployment, lightweight and support lean management and improve the warehouse intelligent upgrading of operation safety.

Description

Technical Field

[0001] This invention relates to a method for upgrading and transforming a logistics warehouse, and more particularly to an intelligent control method for a logistics warehouse based on wireless signal positioning. Background Technology

[0002] Given the current trend of industrial upgrading, most warehouses in China have a need to improve efficiency, save costs, implement lean management, and ensure safety. However, the commonly used methods in the logistics industry, such as using robots, ground sensing chips, or installing smart shelves, require high-cost infrastructure. Without a substantial increase in warehouse operating profit margins, these methods lack economic value and feasibility.

[0003] The main disadvantages of existing conventional logistics warehouses are as follows:

[0004] 1. There are no shelves in the warehouse; goods are stacked on pallets.

[0005] 2. Goods are moved in and out of the warehouse and placed using forklift arms on pallets;

[0006] 3. Forklifts place goods arbitrarily within the warehouse, with forklift drivers placing them based on subjective judgment.

[0007] For the reasons mentioned above, this invention proposes a method for intelligent upgrading and transformation of warehouses that achieves low cost, rapid deployment, lightweight design, and supports lean management and improved operational safety. Summary of the Invention

[0008] The technical problem to be solved by the present invention is to provide an intelligent control method for logistics warehouses based on wireless signal positioning, which can achieve precise management of the loading and unloading position of goods without the need for ground construction or installation of shelves.

[0009] The technical solution adopted by the present invention to solve the above-mentioned technical problems is to provide an intelligent control method for logistics warehouses based on wireless signal positioning, including the following steps: S1) Install a signal receiving base station on the top of the warehouse and install signal transmitting tags on forklifts and cargo pallets, wherein the signal receiving base station is connected to a computer via a network cable; S2) The signal transmitting tags periodically broadcast acceleration status to the signal receiving base station for positioning and behavior notification; S3) The computer receives concurrent data queues from the signal receiving base station to determine the cargo loading and unloading process and cargo location.

[0010] Furthermore, in step S1, two sensing and positioning tags are installed on the forklift, one on the top of the forklift and the other on one side of the forklift arm; at the same time, a sensing and positioning tag is installed on one side of the pallet, and when the forklift is loading goods, the sensing and positioning tags on the forklift arm and the pallet are on the same side.

[0011] Furthermore, in step S2, the signal transmitting tag senses motion using a three-axis or higher accelerometer and broadcasts the acceleration status to the signal receiving base station at a frequency of 3 to 5 Hz during motion.

[0012] Furthermore, in step S2, each acceleration state transmitted by the signal transmitting tag carries a unique signal logic identifier, and the computer filters invalid reflection data from the tag using the signal logic identifier.

[0013] Further, step S3 includes: S31) The computer calculates the position of each tag signal using a positioning algorithm; S32) The acceleration states of the pallet tag where the goods are located and the tag on the forklift arm are obtained, and the similarity of the tag movement behavior is used to determine whether the goods are loaded on the forklift; S33) If the goods are loaded on the forklift, the position of the signal tag on the top of the forklift is taken as the position of the goods, and its movement trajectory is recorded in a memory mapping table with the ID of the pallet tag where the goods are located as the key value; S34) After the forklift places the goods to the destination, the data of the stationary state generated by the acceleration state of the tag on the pallet where the goods are located is used to determine whether the goods have been unloaded and placed; S35) The goods placement position is obtained by looking up its position trajectory data in the memory mapping table with the ID of the pallet tag where the goods are located as the key value; S36) The goods placement position is matched with the logical storage location number in the warehouse, and the placement status is fed back to the warehouse management system to realize seamless intelligent management of goods in the warehouse.

[0014] Furthermore, the process for judging the similarity of tag motion behavior in step S32 is as follows: acquire the multi-axis acceleration data of all tags collected at the same time and calculate the unit vector; then project it onto the same plane, and then use FFT Fourier transform on the two-dimensional waveform to analyze the time domain and frequency domain, and use the relation coefficient and divergence comparison to identify whether they are similar waveforms; if the multi-axis motion acceleration waveform of the pallet tag where a certain goods is located is similar to that of the tag on the forklift arm, then it is determined that the goods are loaded on the forklift.

[0015] Furthermore, the process for judging the similarity of the label movement behavior in step S32 is as follows: using the pallet label to sense the acceleration state in multiple axes, if there is acceleration in any direction within a preset duration, then it is determined that the pallet label is moving; for a pallet label in motion, if the forklift arm position and the cargo pallet position are simultaneously within a certain range, and the cargo pallet vibration duration reaches a preset duration, then it is determined that the cargo is on the forklift.

[0016] Furthermore, the process for determining whether the goods have been unloaded and placed in step S34 is as follows: For a pallet label in a loading state, if the data duration of the acceleration state generating the static state exceeds a preset threshold, it is determined that the goods have been unloaded and placed.

[0017] Furthermore, the process of obtaining the goods placement location in step S35 is as follows: After determining the goods placement, return to the earliest moment when static data was generated, search for the goods location set in the mapping table with the goods pallet label number as the key value, determine the placement location column using the density statistics method according to the warehouse location rules, calculate and obtain the placement location based on the warehouse location management rules, and then accumulate the difference between the forklift top label and the center position of the goods as the goods placement location.

[0018] Compared with existing technologies, this invention offers the following advantages: The intelligent control method for logistics warehouses based on wireless signal positioning provided by this invention, by installing signal transmitting tags on forklifts and cargo pallets, and installing suction cup-type signal receiving base stations on the warehouse roof, achieves high-precision management of cargo loading and unloading status and placement location through wireless positioning and motion sensing. This enables a low-cost, rapid deployment, lightweight approach to warehouse intelligent upgrades that supports lean management and improves operational safety. Based on previous research and implementation experience, intelligent warehouse upgrades can be deployed in a 3000-square-meter warehouse within 48 hours, and even with full warehouse capacity, the cost can be controlled to within 5% of the cost of using ground-sensor chips or intelligent shelving systems. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the intelligent control system architecture for logistics warehouses based on wireless signal positioning according to the present invention;

[0020] Figure 2 This is a flowchart illustrating the intelligent control process for logistics warehouses based on wireless signal positioning according to the present invention.

[0021] Figure 3 This is a loading and judgment logic diagram for the present invention;

[0022] Figure 4 This is the unloading judgment logic diagram of the present invention;

[0023] Figure 5 The present invention provides a logic diagram for calculating the location of goods. Detailed Implementation

[0024] The present invention will now be further described with reference to the accompanying drawings and embodiments.

[0025] Figure 1 This is a schematic diagram of the intelligent control system architecture for logistics warehouses based on wireless signal positioning according to the present invention.

[0026] Please see Figure 1The intelligent control system for logistics warehouses based on wireless signal positioning provided by the present invention consists of a signal receiving base station 1 installed on the top of the warehouse, signal transmitting tags 2 installed on forklifts and cargo pallets, and a computer 3. The positioning and cargo management algorithm model is installed in the computer 3, and the base station 1 and the computer 3 are connected by a network cable via a switch 4.

[0027] The main components and uses of this invention are described below:

[0028] Signal receiving base station 1: Primarily used for locating and sensing the position of signal transmitting tags. A mature AOA (Angle-of-Arrival) positioning base station can be selected. Generally, AOA positioning, in the absence of signal obstruction or metal interference, can achieve a positioning accuracy of 10cm. AOA base stations use phased array antennas to determine location based on the signal's angle of incidence. AOA is currently the most accurate known method for commonly used wireless positioning. The effective signal reception range of each base station can be evenly configured according to the installation height in the warehouse, with the coverage area being a circle with the height data as the radius.

[0029] Sensing and positioning tag 2: It senses behavior by using a three-axis or higher accelerometer and broadcasts the acceleration status to the signal receiving base station at a frequency of 3 to 5 Hz when moving, for positioning and behavior notification.

[0030] Computer 3: Hardware requirements: 16GB or more of memory; 4-16 core CPU selected based on the number of base stations. Software mainly consists of a high-concurrency data queue for receiving base station data, a cargo loading / unloading algorithm model, and a cargo location estimation algorithm model.

[0031] Installation instructions and improvements for the sensing and positioning signal tag 2:

[0032] 1) Install two sensing and positioning tags on the forklift, one on the top of the forklift and the other on one side of the forklift arm.

[0033] 2) Fix the tag to one side of the pallet. At least one sensing and positioning tag should be installed on the same side of the forklift arm when the forklift is loading goods, in order to meet the requirements of the low power consumption and low cost goods loading and unloading algorithm model. See the algorithm model below.

[0034] The reason for the above improvements is that directly using the industry-standard AOA (Optical Oriented Array) positioning method can achieve a positioning accuracy of 10cm in an interference-free, open, unobstructed area with no signal blockage and signals transmitted at the same plane height. However, in practical applications, especially in complex environments such as warehouses, the accuracy falls far short of the theoretical level due to factors such as goods obstructing the signal, the absorption of waves by the materials, and the varying heights of the goods. Based on previous research and experiments using different frequency bands of radio waves, in warehouses with stacked goods, when signal transmission tags were placed directly on the goods, the positioning error was mostly within the range of 1 to 13 meters and was irregular. The positioning data was therefore worthless and could not be directly applied. Hence, the above modification method is necessary.

[0035] For the reasons mentioned above, this invention needs to overcome signal obstruction to ensure that signals are transmitted on the same height plane and to achieve high-precision positioning. Since forklifts are required for lifting and lowering goods in a warehouse, this invention uses a forklift-based indirect positioning method.

[0036] 1) Installing signal transmitting tags on the top of the forklift will ensure unobstructed access and maintain a level plane.

[0037] 2) Install signal transmitting tags on the forklift arm and the cargo pallet. Since the forklift arm and the cargo pallet are basically at the same horizontal height and have similar motion characteristics, they can be used to analyze the relationship between the bound cargo and the forklift loading and unloading (the loading and unloading algorithm model is below) so as to realize the indirect positioning of cargo by the forklift and make the goal of refined management of cargo loading and unloading possible.

[0038] Please see Figure 2 The main data flow of this invention is as follows:

[0039] 1) When the positioning tag 2 is in motion, the accelerometer broadcasts multi-axis acceleration data and positioning signal data at a frequency of 3 to 5 times per second. The base station that receives the signal sends the data to the computer data queue through the network.

[0040] 2) At the same time, computer 3 calculates the position of each tag signal using positioning algorithms such as AOA.

[0041] 3) By combining the multi-axis acceleration data and position data of the pallet label containing the goods with the signal label on the forklift arm, and using the loading model algorithm, determine whether the goods are loaded on the forklift.

[0042] 4) If the goods are loaded on the forklift, the position of the signal tag on the top of the forklift is used as the position of the goods. This is managed by memory mapping with the goods pallet tag ID as the key value, and its movement trajectory is recorded.

[0043] 5) After the forklift places the goods to the destination, the acceleration data of the label on the pallet containing the goods will generate static data. At this time, the unloading algorithm model will determine whether the goods have been placed.

[0044] 6) Use the memory-mapped table with the cargo pallet label ID as the key to find its location trajectory data, and analyze the cargo placement location through the cargo location inference model.

[0045] 7) Match the location with the logical storage location number in the warehouse, and feed back the placement status to the warehouse management system to achieve seamless and intelligent management of goods in the warehouse.

[0046] It is important to note that the acceleration and position data transmitted by the tag must be accompanied by a signal logic identifier for each transmission. The computer uses this identifier to filter out invalid reflection data (ambient echo) from the tag.

[0047] The loading and unloading model algorithm of this invention is divided into two parts: a loading model and an unloading model. Each model can be independently configured with its running step size and cycle (duration and frequency of the cycle) according to the application scenario. Based on the experiment and the speed and rhythm of warehouse forklifts, a cycle of no more than once every 0.5 seconds is suitable. The loading model is mainly used to analyze whether goods are loaded on a forklift and on which forklift. The unloading model is mainly used to analyze whether goods are placed down by the forklift and whether they are unloaded into a suitable storage location.

[0048] Loading Models: Based on the hardware design using sensing tags, two loading models are proposed: one uses multi-axis sensing to analyze the motion acceleration vector curve, and the other directly analyzes the existence of acceleration from multi-axis sensing. Both models have met functional requirements in preliminary tests and are suitable for refined operations in ordinary warehouses.

[0049] 1) Motion acceleration vector curve method:

[0050] By utilizing the waveform similarity of the frequency, direction, and magnitude of multi-axis motion acceleration sensed by tags, we can analyze and determine the similarity of the motion behavior between the cargo pallet tag and the forklift arm tag during forklift loading, thus determining whether the cargo has been loaded onto the forklift. The main analysis method is as follows.

[0051] Calculate the unit vector based on the multi-axis acceleration data of all tags collected at the same time:

[0052] iOV = OV / |OV|;

[0053] ih = iVW^iOV;

[0054] iv=iOV^ih;

[0055] Due to the different ways and locations of the label installation, the coordinate systems of the vectors are inconsistent. To simplify the projection onto the same plane, it is used for waveform similarity analysis in a two-dimensional coordinate system.

[0056]

[0057]

[0058]

[0059] Then, the FFT Fourier transform method is used to analyze the time domain and frequency domain of the two-dimensional waveform. The relationship coefficients and divergence are compared to identify whether they are similar waveforms.

[0060] KL divergence:

[0061]

[0062] JS divergence:

[0063]

[0064] This invention utilizes a three-axis accelerometer to simplify the projection to a two-dimensional waveform, and then identifies similarity to satisfy the judgment of warehouse forklift behavior. In scenarios with fine-grained motion behavior, accelerometers with more than three axes can be used to compare with multi-dimensional waveforms, and many mathematical methods can be employed. This invention only provides a method of judging forklift loading behavior by analyzing the waveform of motion acceleration, and is not limited to specific mathematical equations.

[0065] Features of the motion acceleration vector curve method: high accuracy in loading judgment, with an accuracy rate of up to 99.9% according to experimental loading analysis; high requirements for tag hardware, slightly higher price, and higher power consumption; suitable for scenarios that can accommodate larger batteries.

[0066] 2) Method for determining the existence of motion acceleration:

[0067] The model utilizes a multi-axis sensing method to detect acceleration in any direction using pallet tags to determine if the tags are moving. It also combines the simultaneous presence of the forklift arm position and the pallet position within a certain range with the duration of pallet vibration to determine if the goods are on the forklift. The model logic is as follows: Figure 3 As shown. Vibration is the sensing of acceleration on any one axis of a multi-axis acceleration system. Because the ground generally shakes when a heavy object passes over it, the goods on the ground also shake. This is used to filter out this kind of pre-emptive vibration that is not caused by the loading and unloading of goods.

[0068] It should be noted that the tag acceleration value is read at the time of signal transmission. Even in motion, there may be cases where there is no data for any of the three axes, such as the state 00110001110010101. In such cases, signal time filtering is required.

[0069] Positioning depends on different algorithms. Due to inconsistent signal transmission times and limitations in computing power, signals that are physically close at the same time may have time delays in the calculation results, leading to large position differences. Therefore, the model uses periodic detection to determine the relationship between goods and forklifts.

[0070] The characteristics of multi-axis sensing acceleration are: the tag hardware requirements are the lowest, the power consumption is the least, and the same battery can last at least twice as long; the accuracy is slightly lower, but according to more than a thousand tests, it can reach 93.6%. By adding other auxiliary rule logic, it can also have the ability to accurately position the goods.

[0071] Unloading Model: When a forklift is loaded with goods, after the pallet of goods enters a stationary state for 30 seconds, it is determined that the goods have been unloaded and placed. The logic of the unloading model is as follows: Figure 4 As shown.

[0072] Cargo Location Deduction Model: After cargo placement is determined, the regression starts from the earliest static data point. It searches the cargo location set in a mapping table with the cargo pallet label number as the key. Based on warehouse location rules, it uses density statistics to determine the placement location column and calculates the placement location based on warehouse location management rules. Finally, it sums the forklift specifications (the difference between the forklift top label and the cargo center position) to determine the cargo placement location. The model logic is as follows: Figure 5 As shown.

[0073] It should be noted that during the placement of goods by a forklift, the forklift arm needs to be raised. Since the forklift arm is usually made of metal, it is highly likely to affect the position signal, causing the positioning to drift. Therefore, density statistics and storage location rules are needed to help correct the placement position of the goods.

[0074] Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications and improvements without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be defined by the claims.

Claims

1.A logistics warehouse intelligent control method based on wireless signal positioning, characterized in that, Includes the following steps: S1) Install a signal receiving base station on the top of the warehouse and install signal transmitting tags on forklifts and cargo pallets. The signal receiving base station is connected to a computer via a network. S2) The signal transmitting tag periodically broadcasts the acceleration status to the signal receiving base station for positioning and behavior notification; S3) The computer receives a concurrent data queue from the signal receiving base station to determine the cargo loading and unloading process and cargo location; In step S1, two sensing and positioning tags are installed on the forklift, one on the top of the forklift and the other on one side of the forklift arm; at the same time, a sensing and positioning tag is installed on one side of the pallet, and when the forklift is loading goods, the sensing and positioning tags on the forklift arm and the pallet are on the same side. In step S2, the signal transmitting tag senses motion by using a three-axis or higher accelerometer and broadcasts the acceleration status to the signal receiving base station at a frequency of 3 to 5 Hz during motion. Step S3 includes: S31) The computer calculates the location of each tag signal using a positioning algorithm; S32) Obtain the acceleration status of the pallet label where the goods are located and the label on the forklift arm, and determine whether the goods are loaded on the forklift by the similarity of the label movement behavior; S33) If the goods are loaded on the forklift, the position of the signal tag on the top of the forklift is taken as the position of the goods, and its movement trajectory is recorded in the memory mapping table with the pallet tag ID of the goods as the key value. S34) After the forklift places the goods to the destination, it uses the acceleration status of the label on the pallet containing the goods to generate data on the stationary state, and determines whether the goods have been unloaded and placed. S35) In a memory-mapped table with the pallet label ID of the goods as the key, the location of the goods is obtained by looking up its location trajectory data; S36) The location of the goods is matched with the logical storage location number in the warehouse, and the placement status is fed back to the warehouse management system to realize seamless intelligent management of goods in the warehouse; The process of determining the similarity of tag motion behavior in step S32 is as follows: The system uses a tray label to sense acceleration in multiple directions. If acceleration occurs in any direction within a preset duration, the system determines that the tray label is moving. For a pallet label in motion, if the forklift arm position and the pallet position are both within a certain range, and the pallet vibration duration reaches a preset duration, then it is determined that the goods are on the forklift. The process for determining whether the goods have been unloaded and placed in step S34 is as follows: For a pallet label that is in a loading state, if the duration of the data from the acceleration state to the stationary state exceeds a preset threshold, it is determined that the goods have been unloaded and placed. The process of obtaining the cargo placement location in step S35 is as follows: After determining the placement of goods, we return to the earliest moment when static data was generated. In the mapping table with the goods pallet label number as the key value, we search for the goods location set. According to the warehouse location rules, we use the density statistics method to determine the placement location column and calculate the placement location based on the location management rules. Then, we accumulate the difference between the forklift top label and the center position of the goods as the goods placement location. 2.The intelligent control method of a logistics warehouse based on wireless signal positioning according to claim 1, wherein, The signal sending tag in the step S2 sends the acceleration state with a unique signal logic identification each time, and the computer filters the invalid reflection data of the tag through the signal logic identification.

Images