Progressive domain adaptive bionic positioning method

Abstract

Three key indexes of the excellent positioning method are low cost, high precision and low power consumption. However, no technology can simultaneously meet the three requirements at present. The invention discloses a progressive domain adaptive bionic positioning method, and the method comprises the steps: S1, starting rough absolute positioning of a target, and obtaining a rough absolute position of the target; s2, the main body starts absolute positioning, and the absolute position of the main body is obtained; s3, calculating an approximate relative position of the subject and the target according to the rough absolute position of the target and the absolute position of the subject; and S4, when the approximate relative position of the main body and the target is smaller than a threshold value, fine relative positioning is started, and the fine relative position of the main body and the target is obtained. According to the method, different positioning methods are started through a rough relative distance, different positioning technologies are fused, an absolute position and a relative position are ingeniously switched, and a reasonable balance point among low cost, high precision and low power consumption is attempted to be found.

Description

Technical Field

[0001] The present invention relates to positioning technology, and particularly to a progressive domain adaptive bionic positioning method. Background Art

[0002] There are different classification methods for positioning technology. Two classification methods are introduced below.

[0003] Positioning technology can be divided into rough positioning and fine positioning according to accuracy. Rough positioning generally has low cost, low power consumption, and is relatively convenient to implement. Fine positioning generally has high cost, high power consumption, and is troublesome to implement. According to the current development level of positioning technology, an accuracy of one meter is considered rough positioning, and an accuracy of one decimeter is considered fine positioning.

[0004] Positioning technology can be divided into absolute positioning and relative positioning according to different reference points. Absolute positioning uses a unified fixed point as the origin, and all positions will refer to this origin to represent geographical relationships. Relative positioning uses one of the mutually referenced parties as the origin to express the mutual geographical relationships between the two referenced parties.

[0005] The above two methods can be cross-integrated. For example, rough positioning can be further divided into rough absolute positioning and rough relative positioning. Fine positioning can be further divided into fine absolute positioning and fine relative positioning.

[0006] The main criteria for evaluating the advantages and disadvantages of positioning technology are three points: low cost, high accuracy, and low power consumption. There are many positioning technologies, each with its own advantages and disadvantages, and no single method can meet all three of these requirements simultaneously. Taking satellite positioning, which is the most commonly used for positioning, as an example, the global satellite navigation system has low client cost and low power consumption, but the positioning accuracy is about 5 meters, which belongs to rough positioning. In the field of indoor positioning, due to weak satellite signals, many alternative technologies to satellites have emerged, such as wifi, Bluetooth, uwb, lasers, cameras, etc. Whether it is indoor or outdoor positioning, the trade-off between cost, accuracy, and power consumption is faced. Summary of the Invention

[0007] To solve the problems existing in the background technology, the present invention proposes a progressive domain adaptive bionic positioning method. The human brain has excellent positioning capabilities. According to research findings, the nervous systems used by humans for positioning distant and nearby objects are different. When positioning a distant target, people only need to roughly know which area and direction the object is in. At this time, we don't need to concentrate too much attention and don't need to mobilize too many nervous systems. When a target is right in front of us and we need to approach, pick up or follow it, people need to know the orientation more precisely, specifically the orientation in the front, back, left, right, up and down relative to themselves and the distance from themselves. The brain will then activate more complex neurons and require more concentrated attention. Based on this, the present invention proposes a bionic brain positioning method that automatically enables different positioning systems according to the distance of the target.

[0008] The present invention provides a progressive domain adaptive bionic positioning method, characterized in that the method includes: S1, the target starts rough absolute positioning to obtain the rough absolute position of the target; S2, the main body starts absolute positioning to obtain the absolute position of the main body; S3, calculates the approximate relative position between the main body and the target according to the rough absolute position of the target and the absolute position of the main body; S4, when the approximate relative position between the main body and the target is less than the threshold, starts fine relative positioning to obtain the fine relative position between the main body and the target.

[0009] Further, for the step S1 where the target starts rough absolute positioning, the technologies adopted include but are not limited to fingerprint positioning method, GNSS positioning method.

[0010] Further, for the step S2 where the main body starts absolute positioning, the method can be either rough absolute positioning or fine absolute positioning.

[0011] Further, when starting fine relative positioning in step S4, the target can choose to turn off or turn on rough absolute positioning.

[0012] Further, when starting fine relative positioning in step S4, the main body can choose to turn off or turn on absolute positioning. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 is a schematic flow chart of a progressive domain adaptive bionic positioning method provided by the present invention Figure 2 is another schematic flow chart of a progressive domain adaptive bionic positioning method provided by the present invention SPECIFIC IMPLEMENTATION METHOD

[0014] The described examples are only a part of the embodiments in the present disclosure. All other embodiments obtained by those skilled in the art based on the embodiments in the present disclosure without creative efforts belong to the scope protected by the present disclosure.

[0015] As Figure 1 shown, the present invention provides a progressive domain adaptive bionic positioning method, which is characterized in that the method includes: S1, the target starts rough absolute positioning to obtain the rough absolute position of the target; S2, the main body starts absolute positioning to obtain the absolute position of the main body; S3, calculate the approximate relative distance between the main body and the target according to the rough absolute position of the target and the absolute position of the main body; S4, when the approximate relative distance between the main body and the target is less than the threshold, start fine relative positioning to obtain the fine relative position of the main body and the target.

[0016] In step S1 where the target starts rough absolute positioning, the technologies adopted include but are not limited to fingerprint positioning method and satellite positioning method. It mainly depends on the application environment being indoor or outdoor. If it is outdoor, GNSS positioning can be used. GNSS is the abbreviation of Global Navigation Satellite System, that is, the Global Navigation Satellite System. There are four major positioning systems in the world, GPS of the United States, GLONASS of Russia, Galileo of Europe, and Beidou of China. The positioning accuracy of GNSS is generally about 5 meters, belonging to rough positioning. All positioning points of GNSS positioning are represented by the same earth longitude and latitude standard, belonging to absolute positioning. If it is indoor, the fingerprint positioning method can be used. Fingerprint positioning is to select some positions on a known map, associate the fingerprints of these positions with the position coordinates, collect them into a fingerprint database, and then match the real-time obtained fingerprint with the fingerprint database during positioning to estimate the real-time position. According to the source of the fingerprint, fingerprint positioning has wifi, Bluetooth, RFID, geomagnetism, etc. For different fingerprint sources of the fingerprint positioning method, the positioning accuracy generally ranges from 2 meters to 50 meters, belonging to rough positioning. For the fingerprint positioning method, all positioning points are represented on a known map, belonging to absolute positioning.

[0017] In step S2, the main body enables absolute positioning to obtain the absolute position of the main body. The method used can be either rough absolute positioning or fine absolute positioning. Rough absolute positioning, as described above, includes but is not limited to fingerprint positioning and satellite positioning. Fine absolute positioning includes but is not limited to TDOA of UWB, AOA of Bluetooth, and SLAM. TDOA of UWB (short for Time Difference of Arrival, that is, the time difference of arrival) and AOA of Bluetooth (Angle of Arrival) both require deploying base stations on a known map to determine the position coordinates of the base stations. When positioning, based on the geometric relationship between the terminal and the base stations, the position coordinates of the terminal on the known map are calculated. Therefore, both are absolute positioning. The accuracy of TDOA of UWB is up to 10 cm, and the accuracy of AOA of Bluetooth is 30 cm, both belonging to fine positioning. SLAM is short for Simultaneous Localization and Mapping, that is, simultaneous mapping and positioning. The scanned and matched points are all represented on a map, belonging to absolute positioning. The accuracy of SLAM is up to 2 cm, belonging to fine positioning.

[0018] In step S3, based on the rough absolute position of the target and the absolute position of the main body, the approximate relative distance between the main body and the target is calculated. Since only the approximate relative distance between the main body and the target is required in this step and high accuracy is not demanded, the absolute position of the main body in S2 can be rough positioning or fine positioning.

[0019] In step S4, when the approximate relative distance between the main body and the target is less than the threshold, fine relative positioning is enabled to obtain the fine relative position of the main body and the target. Fine relative positioning includes but is not limited to PDOA of UWB, short for Phase Difference of Arrival, that is, the arrival phase difference. This method first determines the direction and then measures the distance. First, the direction angle between the target and the main body is measured by the AOA method, and then the distance between the target and the main body is measured by TOF (short for Time of Flight), thus obtaining the relative position of the target and the main body. The principle of AOA is that the phase difference generated by the signal passing through the antenna array is used to calculate the signal direction. An alternative method can use AOD (short for Angle of Departure), and the signal source can be UWB or Bluetooth.

[0020] The setting of the threshold needs to be comprehensively considered, including the error variance of the approximate relative distance between the main body and the target and the accurate measurement range of fine relative positioning. Different positioning technologies have different high-precision areas and low-precision areas. For example, TDOA of UWB is more accurate within a range of 20 meters and within plus or minus 60 degrees.

[0021] Such asFigure 2 , step S4, after obtaining the fine relative position of the target and the subject, this is enough for the subject to take a series of actions such as picking up or following the target. If you want to continue to know the fine absolute position of the target, you must add a few more steps. In step S4, the target can choose to turn on or off the rough absolute positioning, and the subject can also choose to turn on or off the absolute positioning. If the subject turns on the fine absolute positioning in step S2, and turns on the fine relative positioning in step S4, and does not turn off the fine absolute positioning of the subject, after obtaining the fine absolute position of the target relative to the subject, the fine absolute position of the target can be obtained based on the fine absolute position of the subject.

[0022] In summary, the present invention proposes a progressive domain adaptive bionic positioning method, which is applicable both outdoors and indoors. It is especially suitable for scenarios where targets are scattered in unspecified areas. It draws on the different neural working principles adopted by the brain positioning system according to the distance from itself. The whole system is relaxed and timely dispatches different positioning technologies, adaptively switches between long-range fuzzy search and short-range precise targeting, relative positioning and absolute positioning, continuous detection and intelligent awakening, which is diverse but not chaotic. This positioning method is of great significance to the development of green economy and artificial intelligence.

Claims

1. A progressive domain adaptive bionic positioning method, characterized in that The progressive domain adaptive bionic positioning method includes: S1, the target starts rough absolute positioning to obtain the rough absolute position of the target; S2, the main body starts absolute positioning to obtain the absolute position of the main body; S3, calculate the approximate relative position between the main body and the target according to the rough absolute position of the target and the absolute position of the main body; S4, when the approximate relative position between the main body and the target is less than the threshold, start fine relative positioning to obtain the fine relative position between the main body and the target.

2. The progressive domain adaptive bionic positioning method according to claim 1, characterized in that For the rough absolute positioning started by the target in step S1, the technologies adopted include but are not limited to fingerprint positioning method and satellite positioning method.

3. The progressive domain adaptive bionic positioning method according to claim 1, characterized in that For the absolute positioning started by the main body in step S2, the method can be either rough absolute positioning or fine absolute positioning.

4. The progressive domain adaptive bionic positioning method according to claim 1, characterized in that When starting fine relative positioning in step S4, the target can choose to turn off or turn on the rough absolute positioning.

5. The progressive domain adaptive bionic positioning method according to claim 1, wherein When starting fine relative positioning in step S4, the main body can choose to turn off or turn on the absolute positioning.

Images