Optimization method for sensing target object in communication sensing integration

Abstract

The invention belongs to the technical field of communication and sensing integration, and particularly relates to an optimization method for sensing a target object in communication and sensing integration, and the method comprises the steps: collecting radar sensing data in a set time period, and arranging the radar sensing data; obtaining the moving direction of the target object at the current time according to the starting point position before the time threshold of the target object at the current time and the position of the target object at the current time, sliding the time window according to the time threshold, and calculating the moving direction of the next time point; and / or calculating the optimized position of the target object according to the position of the target object at a certain time and the positions of a plurality of adjacent reference points before and after the certain time; outputting the moving direction of the next time point as an optimized moving direction; and / or outputting optimization positions corresponding to different time points, and connecting the optimization positions according to a time sequence to serve as an optimized moving path. The method has the beneficial effects that the moving direction and the moving path of the target object are optimized in communication and perception integration by taking adjacent perception data before and after a certain moment as reference, and the accuracy of the communication and perception integration is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of synaesthesia integration, and in particular to an optimization method for perceiving a target object. Background Art

[0002] With the development of communication technology, the integration of communication and perception of 5G-A / 6G (5G Advanced, 5G enhanced) has become an important trend. It realizes the unified design of communication and perception functions through joint design of air interface and protocol, reuse of time-frequency-space resources, and sharing of hardware equipment. While enabling wireless networks to communicate and interact, it can also achieve high-precision and refined perception functions, which can be applied to many fields such as waterway ships and vehicle-road collaboration, and can be combined with REDCAP (Reduced Capability) technology, terminals, and networks.

[0003] In the practice of communication and perception integration, perception radar, as an important perception device, outputs target angle data and longitude and latitude data with poor accuracy, resulting in many abnormal conditions when the system presents information related to the target's location, such as large jumps in the target's moving direction and left and right jumps in the target's moving path on both sides of the actual route. Summary of the invention

[0004] The purpose of the present invention is to provide an optimization method for sensing a target object in communication sensing integration to solve the above technical problems;

[0005] An optimization method for sensing a target object in communication sensing integration, comprising:

[0006] Step S1, collecting radar perception data of the target object within a set time period, and arranging the radar perception data in chronological order to obtain serialized data;

[0007] Step S2 comprises,

[0008] Obtaining the moving direction of the target object at the current time based on the starting position before the time threshold of the target object at the current time and the position at the current time, and sliding the time window with the time threshold to calculate the moving direction of the target object at the next time point; and / or,

[0009] Calculating the optimal position of the target object based on the position of the target object at a certain time and the positions of a number of adjacent reference points before and after the certain time;

[0010] Step S3, outputting the moving direction of the target object at the next time point as the optimized moving direction; and / or,

[0011] The optimized positions of the target object at different time points are output and connected in chronological order as the optimized moving path.

[0012] Preferably, the radar sensing data in step S1 includes angle data and longitude and latitude data of the target object.

[0013] Preferably, the moving direction of the target object at the current time in step S2 is obtained by connecting a line from the starting point to the end point, with the position corresponding to the starting time as the starting point and the position corresponding to the current time as the end point;

[0014] The start time is calculated by subtracting the time threshold from the current time.

[0015] Preferably, in step S2, the time window of the serialized data is slid using the time threshold to calculate the moving direction of the target at the next time point.

[0016] Preferably, in step S2,

[0017] An optimization equation is listed based on the spatial position data of the reference point position, and the optimization equation is converted into an optimization matrix by setting boundary conditions. The weight coefficient of each position point is calculated through the optimization matrix to form a weight matrix, and the optimized position of the target object is calculated based on the weight matrix.

[0018] Preferably, the optimization equation is expressed by the following formula:

[0019] Θ(t)=c0+c1t+c2t 2 +c3t 3 +…+cnt n ;

[0020] Wherein, n represents the total number of the position at a certain time and the reference point position;

[0021] t represents the time corresponding to the spatial position of the target object;

[0022] c0, c1, ..., cn represent the weight coefficients.

[0023] Preferably, the optimization matrix is:

[0024]

[0025] Among them, Θ0 represents the initial position at t = 0;

[0026] Θ 0′ represents the initial velocity at t = 0;

[0027] Θ 0″ represents the initial acceleration at t = 0;

[0028] Θ1 represents the first intermediate point position;

[0029] θ2 represents the second intermediate point position;

[0030] Θ m It means t=t m Target position at time

[0031] Θ m′ It means t=t m Target speed at ;

[0032] Θ m″ It means t=t m Target acceleration at ;

[0033] n represents the total number of the positions at a certain time and the positions of the reference points;

[0034] t represents the time corresponding to the spatial position of the target object;

[0035] c0, c1, ..., c(n-1) represent the weight coefficients.

[0036] Preferably, the boundary conditions include the initial position, the initial velocity, the initial acceleration, the first intermediate point position, the second intermediate point position, the target position, the target velocity and the target acceleration.

[0037] Preferably, the weight matrix is ​​also obtained by converting the optimization matrix into a linear algebraic expression and solving it.

[0038] Preferably, the weight matrix is ​​obtained by using the following formula:

[0039] b=(A τ A) -1 A τ X;

[0040] Wherein, b represents the weight matrix;

[0041] A represents the coefficient matrix;

[0042] Z represents the boundary condition matrix;

[0043] A τ Represents the transposed matrix of A.

[0044] The beneficial effect of the present invention is: using adjacent perception data before and after a certain moment as a reference, optimizing the moving direction and moving path of the target object in the communication perception integration, and improving the accuracy of the communication perception integration. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1It is a step diagram of the optimization method of sensing the target object in the communication sensing integration of the present invention. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0048] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments, but they are not intended to limit the present invention.

[0049] An optimization method for sensing target objects in communication and sensing integration, such as Figure 1 As shown, including

[0050] Step S1, collecting radar perception data of the target object within a set time period, and arranging the radar perception data in chronological order to obtain serialized data;

[0051] Step S2 comprises,

[0052] The moving direction of the target at the current time is obtained according to the starting position before the time threshold of the target at the current time and the position at the current time, and the moving direction of the target at the next time point is calculated by sliding the time window with the time threshold;

[0053] and / or,

[0054] Calculate the optimal position of the target object based on the position of the target object at a certain time and the positions of several adjacent reference points before and after the certain time;

[0055] Step S3, outputting the moving direction of the target object at the next time point as the optimized moving direction;

[0056] and / or,

[0057] The optimized positions of the target object at different time points are output and connected in chronological order as the optimized movement path.

[0058] Specifically, the present invention provides an optimization method for sensing a target object in communication and perception integration, which uses a time threshold sliding time window to optimize the moving direction at the next time point, utilizes the continuity of the target object's motion, and by continuously updating the time window, can adjust the estimation of its moving direction according to the target object's latest position change, thereby more accurately optimizing the target object's motion direction, collects a number of radar sensing data within a period of time, serializes them according to time, uses the adjacent sensing data before and after a certain moment as a reference, and calculates the optimized position of the target object by associating the target object position at that moment with the adjacent reference points before and after, thereby improving the accuracy of communication and perception integration.

[0059] After the radar perception data is serialized in time order, the data at adjacent time points are correlated. When the time window is slid by the time threshold, the data used for each direction calculation is continuous and correlated.

[0060] For example, the start time is obtained by subtracting the time threshold from the real time, so that the data in the selected time window can reflect the position change trend of the target object during this period of time.

[0061] New data is continuously added in and old data is gradually removed. This dynamic updating process can make full use of the correlation between data and better track the changes in the target's movement direction.

[0062] In a preferred embodiment, the radar sensing data in step S1 includes angle data and longitude and latitude data of the target object.

[0063] Specifically, the telepathic positioning technology is mainly radar positioning, which uses a telepathic base station to emit electromagnetic wave signals and identify and locate objects by receiving the echo signal energy reflected by the target object.

[0064] The integrated positioning technology provides specific perception data such as the target's movement angle and longitude and latitude calculated by radar.

[0065] Communication and perception integration means that two independent functions, wireless communication and wireless perception, are implemented in the same system and are mutually beneficial. Communication is responsible for the transmission and aggregation of information; perception includes target identification, positioning, imaging, detection and tracking.

[0066] Synaesthesia integration integrates the function of radar into the base station, using one set of equipment to achieve communication and perception at the same time, which can effectively reduce costs and improve spectrum utilization.

[0067] In mobile communication systems, general perception beyond positioning, such as general perception of sensor devices such as radar, will be integrated into the communication system to help mobile operators provide new services such as high-precision positioning and tracking, synchronous imaging, map construction and positioning.

[0068] Target objects refer to objects scanned by the integrated perception radar, including but not limited to ships, vehicles and drones.

[0069] In a preferred embodiment, the moving direction of the target object at the current time in step S2 is obtained by connecting a line from the starting point to the end point, with the position corresponding to the starting time as the starting point and the position corresponding to the current time as the end point;

[0070] The start time is calculated by subtracting the time threshold from the current time.

[0071] Specifically, the time threshold is adjustable. This embodiment takes 15 seconds as an example. Within 15 seconds, the position change of the target object can be approximately represented by the direction of the connecting line to represent its moving direction.

[0072] Taking drones as an example, a drone flying in a certain area is tracked and monitored.

[0073] First, the radar system continuously collects the UAV’s radar perception data within a set time period of 60 seconds. These data include the UAV’s angle data and longitude and latitude data at each moment.

[0074] After the collection is completed, the data are arranged in chronological order to obtain serialized data.

[0075] Then, set the time threshold to 15s. When the real time reaches 30s, the calculation start time is:

[0076] 30-15=15s;

[0077] At this time, take the longitude and latitude position corresponding to 15s (longitude 120.005, latitude 30.002) as the starting point, and the longitude and latitude position corresponding to 30s (longitude 120.030, latitude 30.015) as the end point, and connect the starting point to the end point. The direction of this line is determined as the movement direction of the drone at the moment of 30s.

[0078] Next, 15 seconds is used as the time window of the sliding serialized data to optimize the moving direction of the next time point (i.e., 31 seconds). When the time advances to 31 seconds, the new starting time becomes 31-15=16 seconds.

[0079] At this time, the direction of the connection is re-determined according to the longitude and latitude positions corresponding to 16 seconds and the longitude and latitude positions corresponding to 31 seconds, so as to optimize the estimation of the movement direction of the UAV at 31 seconds.

[0080] This process is repeated throughout the entire monitoring process. As time goes by, the UAV's perception of moving direction is continuously optimized through this sliding time window, so that the UAV's flight trajectory and direction changes can be grasped more accurately, providing reliable direction data basis for subsequent flight control, path planning or safety warning operations.

[0081] Compared with the traditional direction estimation method of a single time point or fixed interval, by utilizing serialized radar perception data and sliding time windows, it can better adapt to changes in the motion state of the target and reduce direction estimation errors caused by data incoherence or local interference.

[0082] As time goes by, the time window is continuously updated so that the estimate of the target's direction can always be adjusted based on the latest position information.

[0083] For example, in an autonomous driving scenario, the vehicle can make decisions such as avoidance or following more timely based on real-time optimized directional information of target objects (such as other vehicles or pedestrians), thereby improving driving safety.

[0084] The configurability of the time threshold increases the flexibility of the method. The time threshold can be adjusted according to the different target movement speeds, movement patterns, and accuracy requirements of the application scenario.

[0085] For example, for fast-moving targets, a smaller time threshold can be set to quickly capture their direction changes; for slow-moving targets, the time threshold can be appropriately increased to reduce the amount of calculation while still ensuring the accuracy of direction estimation.

[0086] In a preferred embodiment, in step S2, a time window of the serialized data is slidably processed using a time threshold to calculate the moving direction of the target at the next time point.

[0087] In a preferred embodiment, in step S2,

[0088] The optimization equation is listed based on the spatial position data of the reference point position, and the optimization equation is converted into an optimization matrix by setting boundary conditions. The weight coefficient of each position point is calculated through the optimization matrix to form a weight matrix. The optimized position of the target object is calculated based on the weight matrix.

[0089] Specifically, for a moving target, its position is constantly changing. This method can better track the actual moving path of the target by processing serialized data and optimizing the moving path of the target.

[0090] For example, when a target object moves irregularly or in a complex environment, it can more flexibly adapt to the dynamic changes of the target object and provide more reliable location information.

[0091] To be more specific, we take n path points as an example to calculate the middle point X {n} The curvature of can be approximately calculated by the following two schemes. The accuracy and stability of the two schemes are different. The window period and other parameters can be modified according to the actual situation (different data ranges and windows will bring different sensitivities and stabilities, and the scheme should be adapted to the actual situation), and equation (1) is obtained.

[0092]

[0093] In a preferred embodiment, the optimization equation is expressed by the following formula:

[0094] Θ(t)=c0+c1t+c2t 2 +c3t 3 +…+cnt n ;

[0095] Where n represents the total number of positions and reference point positions at a certain time;

[0096] t represents the time corresponding to the spatial position of the target object;

[0097] c0, c1, …, cn represent weight coefficients.

[0098] Specifically, the number of reference points is adjustable. The following takes 3 adjacent positions before and after a certain moment, a total of 7 radar scanning positions as an example to calculate the X of the middle point at that moment. {n} Optimized location.

[0099] The position of the target object at a certain moment is associated with the positions of several adjacent reference points before and after that moment, and a relationship equation can be established between these points in space.

[0100] For example, the middle position X at this moment {n} and the first 3 adjacent positions X {n-1} , X {n-2} , X {n-3} , and the next three adjacent positions X {n+1} , X {n+2} , X {n+3} There are 7 points in total, and the position data of these 7 points in space can form a set of equations.

[0101] Θ(t)=c0+c1t+c2t 2 +c3t 3 +…+c7t 7 (2) Formula;

[0102] In a preferred embodiment, the optimization matrix is:

[0103]

[0104] Among them, Θ0 represents the initial position at t = 0;

[0105] Θ 0′ represents the initial velocity at t = 0;

[0106] Θ 0″ represents the initial acceleration at t = 0;

[0107] Θ1 represents the first intermediate point position;

[0108] θ2 represents the second intermediate point position;

[0109] Θ m It means t=t m Target position at time

[0110] Θ m′ It means t=t m Target speed at ;

[0111] Θ m″ It means t=t m Target acceleration at ;

[0112] n represents the total number of positions and reference point positions at a certain time;

[0113] t represents the time corresponding to the spatial position of the target object;

[0114] c0, c1, ..., c(n-1) represent weight coefficients.

[0115] In a preferred embodiment, the boundary conditions include an initial position, an initial velocity, an initial acceleration, a first intermediate point position, a second intermediate point position, a target position, a target velocity, and a target acceleration.

[0116] Specifically, the trajectory optimization is performed using the formula and eight boundary conditions are specified, namely,

[0117] (1) Initial position at t = 0:

[0118] Θ0=c0。

[0119] (2) Initial velocity at t = 0:

[0120] Θ 0′ =c1.

[0121] (3) Initial acceleration at t = 0:

[0122] Θ 0″ =2c2.

[0123] (4) First midpoint position:

[0124] θ1=c0+c1t1+c2t1 2 +c3t1 3 +c4t1 4 +c5t1 5 +c6t1 6 +c7t1 7 .

[0125] (5) Second midpoint position:

[0126] θ2=c0+c1t2+c2t2 2 +c3t2 3 +c4t2 4 +c5t2 5 +c6t2 6 +c7t2 7 .

[0127] (6) Target position at t = tm:

[0128] Θ m =c0+c1tm+c2tm 2 +c3tm 3 +c4tm 4 +c5tm 5 +c6tm 6 +c7tm 7 .

[0129] (7) Target speed at t = tm:

[0130] Θ m′ =c1tm+2c2tm+3c3tm 2 +4c4tm 3 +5c5tm 4 +6c6tm 5 +7c7tm 6 .

[0131] (8) The target acceleration at the local time t = tm is known:

[0132] Θ m″ =c1+2c2+6c3tm+12c4tm 2 +20c5tm 3 +30c6tm 4 +42c7tm 5 .

[0133] Write the above 8 expressions into the optimization matrix in step G32,

[0134]

[0135] Solve the matrix to get the weight coefficient of each position point, as shown in formula (4):

[0136]

[0137] Among them, T = t1-t0, h = Θ1-Θ0, coefficient ci,i = 0,…,7.

[0138] Substituting the weight matrix (4) of the position point into (2), the optimized position of the target at that moment can be calculated. The AI ​​learning model supports matrix calculation, and the AI ​​learning models that can be used include ensemble learning models, support vector machine models, and logistic regression models. After calculating the optimized data of the movement path of the radar-perceived target, it can be output to the system to present the specific movement path of the target.

[0139] In a preferred embodiment, the weight matrix is ​​also obtained by converting the optimization matrix into a linear algebraic expression and solving it;

[0140] The weight matrix is ​​obtained by using the following formula:

[0141] b=(A τ A) -1 A τ x;

[0142] Where b represents the weight matrix;

[0143] A represents the coefficient matrix;

[0144] X represents the boundary condition matrix;

[0145] A τ Represents the transposed matrix of A.

[0146] Specifically, the present invention can also be extended to use mathematical calculation methods (linear algebra) to calculate the matrix to obtain the weight matrix of each position point.

[0147] After obtaining the optimized matrix, when solving the linear equations method, the equation is converted into linear algebraic form:

[0148] A×b=X;

[0149] b=(A τ A) -1 A τ X;

[0150] The weight matrix b can be solved through the above linear algebraic operation. The weight matrix b of the position point can be substituted into formula (1) to calculate the optimal position of the target object at that moment.

[0151] In one embodiment, in the perception data generated by the inter-sensory integrated base station, there are abnormalities in the longitude and latitude data of the target object output by the perception radar (there are errors in the longitude and latitude data of the target object output by the perception radar, resulting in the target object movement path presented by the system not being on the actual route, but jumping left and right with the actual route as the central axis). The method of the present invention is used for optimization, and the implementation steps are as follows:

[0152] In the 5G-A / 6G synaesthesia integration, the present invention collects several adjacent positions before and after a moment as reference points, associates the reference points with the target position at that moment, calculates the optimal position of the target, realizes the optimization of the movement path of the perceived target, and corrects the abnormal latitude and longitude data of the target output by the perception radar through the optimization algorithm to make it more consistent with the actual movement path of the target.

[0153] Among them, 5G-A (5G-Advanced) is based on the evolution and enhancement of 5G network in terms of function and coverage. Its key information technologies include communication perception integration, 3CC (3Component Carriers, three carrier aggregation), etc.

[0154] 6G is the sixth-generation mobile communication standard. Key technologies include spatial multiplexing technologies such as integrated communication and perception, and terahertz frequency bands.

[0155] In one embodiment, in the perception data generated by the inter-sensory integrated base station, there is abnormal target angle data output by the perception radar (the target direction data output by the perception radar has a large error, resulting in a large jump in the target movement direction presented by the system.

[0156] For example, if the ship's heading repeatedly jumps from 0° to 90° and then from 90° to 0° within 1 second, the method of the present invention is used for optimization, and the implementation steps are as follows:

[0157] In the 5G-A / 6G synaesthesia integration, the present invention sets a time threshold as a sliding time window, calculates the moving direction of the target at the next time point, and optimizes the moving direction of the perceived target, thereby correcting the abnormal angle data of the target output by the perception radar to make it more consistent with the actual moving direction of the target.

[0158] The present invention can meet the needs of industry applications for integrated positioning services, has the versatility of various mobile communication standards, and is suitable for optimizing the mobility of targets in mobile communication networks of various standards such as 5G-A and 6G.

[0159] It is also suitable for optimizing the movement patterns of targets using positioning technologies such as 5G, 6G, and radar.

[0160] This article uses 5G-A network as an example for calculation, but it is not limited to 5G-A network, and is also applicable to the optimization of mobile forms of various target objects such as 6G.

[0161] The above description is only a preferred embodiment of the present invention, and does not limit the implementation mode and protection scope of the present invention. For those skilled in the art, it should be aware that all solutions obtained by equivalent substitutions and obvious changes made using the description and illustrations of the present invention should be included in the protection scope of the present invention.

Claims

1. An optimization method for sensing a target object in communication sensing integration, characterized in that: include, Step S1, collecting radar perception data of the target object within a set time period, and arranging the radar perception data in chronological order to obtain serialized data; Step S2 comprises, Obtaining the moving direction of the target object at the current time based on the starting position before the time threshold of the target object at the current time and the position at the current time, and sliding the time window with the time threshold to calculate the moving direction of the target object at the next time point; and / or, Calculating the optimal position of the target object based on the position of the target object at a certain time and the positions of a number of adjacent reference points before and after the certain time; Step S3, outputting the moving direction of the target object at the next time point as the optimized moving direction; and / or, The optimized positions of the target object at different time points are output and connected in chronological order as the optimized moving path.

2. The optimization method for sensing target objects in communication sensing integration according to claim 1, characterized in that: The radar sensing data in step S1 includes the angle data and the longitude and latitude data of the target object.

3. The optimization method for sensing target objects in communication sensing integration according to claim 1, characterized in that: The moving direction of the target object at the current time in step S2 is obtained by connecting a line from the starting point to the end point, with the position corresponding to the start time as the starting point and the position corresponding to the current time as the end point; The start time is calculated by subtracting the time threshold from the current time.

4. The optimization method for sensing target objects in communication sensing integration according to claim 1, characterized in that: In step S2, the time window of the serialized data is slid using the time threshold to calculate the moving direction of the target at the next time point.

5. The optimization method for sensing target objects in communication sensing integration according to claim 1, characterized in that: In step S2, An optimization equation is listed based on the spatial position data of the reference point position, and the optimization equation is converted into an optimization matrix by setting boundary conditions. The weight coefficient of each position point is calculated through the optimization matrix to form a weight matrix, and the optimized position of the target object is calculated based on the weight matrix.

6. The optimization method for sensing target objects in communication sensing integration according to claim 5, characterized in that: The optimization equation is expressed by the following formula: Θ(t)=c0+c1t+c2t 2 +c3t 3 +…+count n ; Wherein, n represents the total number of the position at a certain time and the reference point position; t represents the time corresponding to the spatial position of the target object; c0, c1, ..., cn represent the weight coefficients.

7. The optimization method for sensing target objects in communication sensing integration according to claim 5, characterized in that: The optimization matrix is: Among them, Θ0 represents the initial position at t = 0; Θ 0′ represents the initial velocity at t = 0; Θ 0″ represents the initial acceleration at t = 0; Θ1 represents the first intermediate point position; θ2 represents the second intermediate point position; Θ m It means t=t m Target position at time Θ m′ It means t=t m Target speed at ; Θ m″ It means t=t m Target acceleration at ; n represents the total number of the positions at a certain time and the positions of the reference points; t represents the time corresponding to the spatial position of the target object; c0, c1, ..., c(n-1) represent the weight coefficients.

8. The optimization method for sensing target objects in communication sensing integration according to claim 7, characterized in that: The boundary conditions include the initial position, the initial velocity, the initial acceleration, the first intermediate point position, the second intermediate point position, the target position, the target velocity, and the target acceleration.

9. The optimization method for sensing target objects in communication sensing integration according to claim 5, characterized in that: The weight matrix is ​​also obtained by converting the optimization matrix into a linear algebraic expression and solving it.

10. The optimization method for sensing target objects in communication sensing integration according to claim 5, characterized in that: The weight matrix is ​​obtained by using the following formula: b=(A τ A) -1 A τ X; Wherein, b represents the weight matrix; A represents the coefficient matrix; X represents the boundary condition matrix; A τ Represents the transposed matrix of A.

Images