CN116954235A - AGV trolley navigation control method and system - Google Patents

Abstract

The invention discloses a navigation control method and a system of an AGV (automatic guided vehicle). The method comprises the steps of firstly, carrying out fusion calculation on Bluetooth estimated coordinates and geomagnetic estimated coordinates in a decision layer to calculate the position of the AGV, thereby reducing positioning errors and not consuming additional hardware resources; and then repeatedly calculating the shortest linear distance between the current position of the AGV trolley and the position to be reached by the AGV trolley until the position to be reached by the AGV trolley is reached, so that the whole process is simple in algorithm, short in time consumption and low in hardware requirement, and the requirements of practical application can be well met.

Description

AGV car navigation control method and system

Technical field

The present invention relates to the field of navigation technology, in particular to an AGV car navigation control method and system.

Background technique

AGV (Automatic Guided Vehicle) is an unmanned automated transportation equipment that can carry a certain weight and operate autonomously between the departure point and the destination. It is an important component of the automatic logistics system and flexible manufacturing system. It has good market prospects and application value.

AGVs are divided into orbital AGVs and non-orbital AGVs. Orbital AGVs are AGV equipment that moves according to a preset track, while non-orbital AGVs are AGV equipment with autonomous navigation functions. Common navigation methods of existing non-orbital AGVs The method is to use a lidar navigator fixed on it to scan the environmental contours of the space and analyze it to obtain environmental data. Then the AGV navigates and moves based on the environmental data, but only relies on a single lidar in narrow places with few reference points. The navigator scans the contours of the environment, and the environmental data obtained is prone to deviations, causing the AGV to be unable to navigate normally. In addition, since the above-mentioned lidar navigator is fixedly installed, if the lidar navigator is blocked by a high-level obstacle, the laser The radar navigator cannot scan the environmental contours of the space, which also causes the AGV to be unable to navigate normally. If multiple lidar navigators are set up to scan the environmental contours from multiple angles to overcome the above technical problems, it will increase the production cost of the AGV equipment. .

Chinese patent application number CN201510761432.9 discloses a visual AGV navigation system, including: a visual sensor, an image capture card, an image processor, a PC host and a drive system, wherein: the visual sensor communicates with the image capture card through a USB interface The image acquisition card is connected to the image processor, the image processor is connected to the PC host through an RS232 interface, a USB interface and a JTAG interface, and the image processor is connected to the image processor through a PWM output interface. The image processor is connected to the driving system; the image processor includes a filter processing unit, an edge processing unit and a threshold processing unit that are connected in sequence. Although this navigation system can realize automatic navigation of AVG, the entire algorithm is complex and time-consuming.

Therefore, there is an urgent need for a low-cost, stable and reliable positioning and navigation method for non-orbital AGVs.

Contents of the invention

In order to solve the above technical problems, the present invention provides an AGV car navigation control method and system.

In order to achieve the above objects, the present invention is implemented according to the following technical solutions:

The first object of the present invention is to provide an AGV car navigation control method, which includes the following steps:

S1. Draw an indoor coordinate grid diagram and mark the locations and coordinates of all obstacles in the coordinate grid diagram;

S2. Set up several Bluetooth beacons in the indoor area, and install Bluetooth receivers, iBeacon software and Hall sensors on the AGV to receive Bluetooth and geomagnetic signals respectively, and calculate Bluetooth estimated coordinates and geomagnetic estimated coordinates;

S3. Then, the Bluetooth estimated coordinates and the geomagnetic estimated coordinates are fused at the decision-making level to calculate the current position of the AGV car. ;

S4. Obtain the coordinates of the position where the AGV car is to arrive. , and then according to the formula Find the shortest straight-line distance between the current position of the AGV car and the position where the AGV car is to be reached, and control the AGV car to move along the shortest straight-line distance to the position where the AGV car is to be reached;

S5. Collect the road image in front of the AGV car in real time through the visual sensor on the AGV car;

S6. When it is discovered that there is an obstacle directly in front of the AGV car, control the AGV car to move left or right, and collect the image of the road in front of the AGV car in real time until there is no obstacle directly in front of the AGV car; obtain the coordinates of the current position of the AGV car again and calculate Obtain the shortest straight-line distance between the current position of the AGV car and the position to be reached by the AGV car, and control the AGV car to move along the shortest straight-line distance to the position to be reached by the AGV car;

S7. Repeat steps S5-S6 until the AGV car reaches the waiting position of the AGV car.

Further, the step S2 specifically includes: assuming that the Bluetooth estimated coordinates are: ;During this period, the set of geomagnetic estimated coordinates is:/> ;The coordinates after averaging the weighted geomagnetic estimation coordinate set are: ;Assign weights to coordinates/> and b , the final estimated position coordinates after fusion are:/> ;where:/> <b .

Further, the weight ratio of the Bluetooth and geomagnetic positioning results is set to 1: N, that is: ; Finally, the final estimated position coordinates after fusion are written as:/> .

The second object of the present invention is to provide an AGV car navigation control system, which includes setting a number of Bluetooth beacons in an indoor area, an AGV car processor installed on the AGV car, a Bluetooth receiver installed on the AGV car processor, and iBeacon software and Hall sensor, visual sensor and drive system, wherein: the visual sensor is connected to the AGV car processor through a USB interface, the visual sensor is connected to the drive system through a PWM output interface; the AGV car The processor is used to execute the AGV car navigation control method.

Compared with the existing technology, the present invention first calculates the position of the AGV car by fusing the Bluetooth estimated coordinates and the geomagnetic estimated coordinates at the decision-making level, which reduces the positioning error without consuming additional hardware resources; and then repeatedly calculates the position of the AGV car. The shortest straight-line distance between the current position and the AGV car's waiting position, until it reaches the AGV car's waiting position, the whole process has a simple algorithm, short time consumption, low hardware requirements, and can well meet the needs of practical applications.

Description of the drawings

Figure 1 shows the indoor coordinate grid drawn.

Figure 2 is a schematic diagram of the online positioning stage of the indoor positioning algorithm based on location fingerprint.

Figure 3 is the decision-making layer fusion structure diagram.

Detailed ways

In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be further described in detail below in conjunction with the examples. The specific embodiments described here are only used to explain the invention and are not intended to limit the invention.

This embodiment provides an AGV car navigation control system, which includes setting several Bluetooth beacons in an indoor area, an AGV car processor installed on the AGV car, a Bluetooth receiver installed on the AGV car processor, and iBeacon software and Hall Sensor, visual sensor and drive system, wherein: the visual sensor is connected to the AGV car processor through a USB interface, the visual sensor is connected to the drive system through a PWM output interface; the AGV car processor is used to perform the following AGV car navigation Control method to realize the AGV car reaching the position where the AGV car is to arrive.

The above system can be used to realize AGV car navigation control. The specific steps are as follows:

S1. As shown in Figure 1, draw an indoor coordinate grid diagram, and mark the positions and coordinates of all obstacles in the coordinate grid diagram (larger dots in Figure 1 represent obstacles);

Set up several Bluetooth beacons in the indoor area, and install Bluetooth receivers, iBeacon software and Hall sensors on the AGV car to receive Bluetooth and geomagnetic signals respectively, and calculate Bluetooth estimated coordinates and geomagnetic estimated coordinates;

This embodiment builds a Bluetooth and geomagnetic fingerprint database based on the indoor positioning algorithm of location fingerprint, which is divided into an offline database construction stage and an online positioning stage:

(1) Offline database construction stage

In the offline database construction phase, data information is first collected to establish a fingerprint database for each location. That is, a certain number of reference points with moderate spacing are selected for sampling in a designated indoor area. The location fingerprint information of these points includes the received fingerprints of several APs. RSSI sequence and the coordinates of the current location, and then save these fingerprint data in the location fingerprint database.

(2) Online positioning stage

As shown in Figure 2, the online positioning stage needs to use the established fingerprint database to estimate the user's actual location. You can collect the RSSI signals of several APs currently received through the Bluetooth receiver and iBeacon software installed on the AGV car, and then compare them with The RSSI data in the pre-saved location fingerprint database are compared in sequence to obtain the fingerprint data closest to the currently collected RSSI signal, and the actual location of the current AGV car can be estimated.

Then, the K nearest neighbor algorithm is used for matching estimation, and then the Bluetooth estimated coordinates and geomagnetic estimated coordinates are calculated respectively.

The K nearest neighbor algorithm is a commonly used position fingerprint positioning algorithm. It is assumed that there are m sampling points in the positioning space. AP, then the RSSI vector of the point to be located is/> , the RSSI vector of the i- th sampling point is , the position coordinates are/> , the Euclidean distance between two points can be calculated as: ; where rssi _j represents the received signal strength of the j- th AP, and rssi _ij is the strength of the j -th AP in the i -th sampling point. Then sort the m distances obtained, select the K sampling points with the smallest distance, and then take the mean of these coordinates as the estimation result, that is:/> ;Formula/> Represents the coordinates of the i- th sampling point, /> is the corresponding estimated value. In addition to the estimation error, the cumulative distribution function (CDF) can also be used to evaluate the positioning effect./> Indicates that X is less than or equal to/> The probability of all values appearing, that is:/> ; Analyzing the entire positioning process, it can be seen that there are two reasons that will have a greater impact on positioning accuracy: First, the accuracy of fingerprint database construction. The second is the algorithm used in online matching. It is best to meet the accuracy of the matching algorithm without causing excessive hardware overhead. This can not only improve the accuracy of positioning, but also facilitate large-scale promotion.

S2. After that, the Bluetooth estimated coordinates and the geomagnetic estimated coordinates are fused at the decision-making layer to calculate the user's location.

This embodiment uses a precise positioning function called iBeacon launched by Apple in 2013. It uses Bluetooth technology. Electronic devices around it can sense the Bluetooth signal it emits, and then implement it through a combination of software and hardware. Indoor Positioning. The iBeacon signal covers a range of approximately 5080 m. Users can detect iBeacon signals through applications on mobile terminals such as mobile phones. When the iBeacon Bluetooth beacon is working, the device will transmit data packets at regular intervals, and the interval can be divided into 300 ms, 500 ms or 900 ms. A complete iBeacon system should contain one or several iBeacon beacons, which emit unique identification codes within a certain distance, and the iBeacon signal can be found on the corresponding receiving device. The closer the beacon is to the beacon, the stronger the signal will be. As the distance increases, the signal will continue to attenuate. When the distance exceeds a certain value, the signal will not be detected.

This embodiment uses a magnetic sensor principle similar to the "Hall effect" of smartphones. Specifically, the earth itself has a strong magnetic field, and the magnetic field intensity is different at different locations. The magnetic induction intensity in three directions of x, y and: can be detected by the magnetic sensor, where the x and Y directions are parallel to the ground at that point, and the axis is perpendicular to the ground at that point. The Hall sensor installed on the AGV car is used to collect geomagnetic data of three axes, and the three-axis data is converted into a module value. At this time, the module value is relatively stable regardless of the posture of the AGV car.

Assume that the components of the three axes of the magnetometer at a certain point are , then the modulus value of the magnetic field intensity at this point can be obtained as:/> ;The principle is that the voltage on the side of a constant current conductor will change linearly with the change of the external magnetic field. Therefore, as long as the voltage on the side of the energized conductor is measured, the strength of the surrounding geomagnetic field can be estimated.

Information fusion is a technology that combines multiple sensors and allows them to exert their respective advantages. It has certain information complementary capabilities and can obtain data information not collected by a single sensor, which increases the credibility of the judgment. Generally speaking, information can be fused at the decision-making layer, feature layer, or data layer. This embodiment uses decision-making layer fusion, that is, different sensors are used to monitor a measured object. Each sensor independently completes a series of processes such as data preprocessing, feature information extraction, and target identification to obtain an initial analysis of the measured object. , and then fused at the decision-making level to obtain the final result. Compared with the other two fusion technologies, the decision-making layer fusion technology has the advantages of strong flexibility, small amount of data processed in the fusion stage, good real-time performance, and good fault tolerance. The decision-making layer fusion structure diagram is shown in Figure 3.

First, Bluetooth and geomagnetic information are collected respectively, and then their respective fingerprint databases are constructed. On this basis, the K nearest neighbor algorithm is used for matching estimation, and then the geomagnetic and Bluetooth positioning values are calculated respectively, and then the two are weighted to determine the final result. Assume that the Bluetooth estimated coordinates are: ; During this period, the set of geomagnetic estimated coordinates is: ;The coordinates after averaging the weighted geomagnetic estimation coordinate set are: ;Assign weights to coordinates/> and b , the final estimated position coordinates after fusion are:/> ;On the one hand, the weighted values of the two/> and b are affected by Bluetooth and geomagnetic estimation accuracy. On the other hand, because Bluetooth has more beacons, its fingerprint database data is more than the geomagnetic fingerprint database, so that when outputting one Bluetooth estimation result, it will output multiple geomagnetic estimation results, thereby obtaining an estimate of geomagnetic positioning. gather. Because Bluetooth and geomagnetic results appear at different frequencies, their contributions to positioning are also different. Therefore, when fusing their results, the weight ratio of Bluetooth and geomagnetic positioning results is set to 1: N, that is:/> ;Finally, the final estimated position coordinates after fusion are written as:/> .

It should be noted that if the electromagnetic signal interference is small, the accuracy of Bluetooth will be greatly improved, and the weights can also be flexibly adjusted.

Therefore, the current position of the AGV car is written as , the smaller dot in the upper left corner of Figure 1 represents the assumed current position of the AGV car;

S4. Obtain the coordinates of the position where the AGV car is to arrive. (The smaller dot in the lower right corner of Figure 1 represents the position where the AGV car will arrive), and then according to the formula /> Find the shortest straight-line distance between the current position of the AGV car and the position where the AGV car is to be reached, and control the AGV car to move along the shortest straight-line distance to the position where the AGV car is to be reached;

S5. Collect the road image in front of the AGV car in real time through the visual sensor on the AGV car;

S6. When it is discovered that there is an obstacle directly in front of the AGV car, control the AGV car to move left or right, and collect the image of the road in front of the AGV car in real time until there is no obstacle directly in front of the AGV car; obtain the coordinates of the current position of the AGV car again and calculate Obtain the shortest straight-line distance between the current position of the AGV car and the position to be reached by the AGV car, and control the AGV car to move along the shortest straight-line distance to the position to be reached by the AGV car;

S7. Repeat steps S5-S6 until the AGV car reaches the waiting position of the AGV car.

To sum up, this embodiment first calculates the position of the AGV car by fusing the Bluetooth estimated coordinates and the geomagnetic estimated coordinates at the decision-making layer, which reduces the positioning error without consuming additional hardware resources; and then repeatedly calculates the current position of the AGV car. The shortest straight-line distance between the position and the position where the AGV car is to be reached, until it reaches the position where the AGV car is to be reached, the whole process has a simple algorithm, short time consumption, and low hardware requirements.

The technical solution of the present invention is not limited to the above-mentioned specific embodiments. All technical modifications made based on the technical solution of the present invention fall within the protection scope of the present invention.

Claims (5)

1. The AGV trolley navigation control method is characterized by comprising the following steps of:

s1, drawing an indoor coordinate grid diagram, and marking the positions and coordinates of all obstacles in the coordinate grid diagram;

s2, setting a plurality of Bluetooth beacons in an indoor area, installing a Bluetooth receiver, iBeacon software and a Hall sensor on an AGV trolley to respectively receive Bluetooth and geomagnetic signals, and calculating Bluetooth estimated coordinates and geomagnetic estimated coordinates;

s3, fusing the Bluetooth estimated coordinates and the geomagnetic estimated coordinates in a decision layer to calculate the current position of the AGV；

S4, obtaining coordinates of the position to be reached of the AGV trolleyThen->Obtaining the shortest linear distance between the current position of the AGV trolley and the position to be reached of the AGV trolley, and controlling the AGV trolley to move towards the position to be reached of the AGV trolley along the shortest linear distance;

s5, acquiring a road image in front of the AGV in real time through a visual sensor on the AGV;

s6, controlling the AGV to move leftwards or rightwards when an obstacle exists right in front of the AGV, and collecting road images in front of the AGV in real time until no obstacle exists right in front of the AGV; obtaining the coordinates of the current position of the AGV trolley again, obtaining the shortest linear distance between the current position of the AGV trolley and the position to be reached of the AGV trolley, and controlling the AGV trolley to move along the shortest linear distance to the position to be reached of the AGV trolley;

s7, repeating the steps S5-S6 until the AGV trolley reaches the position where the AGV trolley is to be reached.

2. The AGV car navigation control method according to claim 1, wherein the step S1 specifically includes:

s11, respectively acquiring Bluetooth and geomagnetic information of indoor areas, and further constructing respective fingerprint databases;

and S12, performing matching estimation by using a K neighbor algorithm, and then respectively calculating Bluetooth estimated coordinates and geomagnetic estimated coordinates.

3. The AGV car navigation control method according to claim 1, wherein the step S2 specifically includes:

suppose that the bluetooth estimate coordinates are:the method comprises the steps of carrying out a first treatment on the surface of the The geomagnetic estimation coordinate set during this period is:the method comprises the steps of carrying out a first treatment on the surface of the The geomagnetic estimated coordinate set is weighted by average, and the average coordinates are as follows:the method comprises the steps of carrying out a first treatment on the surface of the The coordinates are respectively given weight +.>Andbobtaining a fused final position estimateThe coordinates are: />The method comprises the steps of carrying out a first treatment on the surface of the Wherein: /> <b。

4. The AGV navigation control method according to claim 3 wherein the weight ratio of the Bluetooth and geomagnetic positioning results is set to 1:N, namely:and finally, writing the fused final position estimation coordinates as:。

5. the utility model provides an AGV dolly navigation control system, its characterized in that, include set up a plurality of bluetooth beacons, install AGV dolly treater on the AGV dolly at indoor region, install bluetooth receiver and iBeacon software and hall sensor, vision sensor and actuating system on the AGV dolly treater, wherein: the visual sensor is connected with the AGV trolley processor through a USB interface, and the visual sensor is connected with the driving system through a PWM output interface; the AGV trolley processor is configured to execute the AGV trolley navigation control method according to any one of claims 1-4.