Underground space vehicle-mounted through-the-wall radar path optimization method based on improved ant colony algorithm

Abstract

The invention discloses an underground space vehicle-mounted through-wall radar path optimization method based on an improved ant colony algorithm, and the method comprises the steps: firstly, considering the particularity of an underground space, designing a self-adaptive rule of a heuristic function according to the distance of a motion path of an unmanned vehicle and the imaging quality of a three-dimensional through-wall radar, and balancing the global search performance and convergence speed of an algorithm; secondly, by solving a non-convex optimization problem under multiple constraint conditions including path length, imaging quality, Gini impurity and the like, it is ensured that environment perception, accurate imaging and path planning are efficiently and accurately completed in a complex underground environment. The method can be applied to underground space unmanned detection and target search and rescue, and has high adaptability and high efficiency.

Description

Technical Field

[0001] The present invention belongs to the field of underground space unmanned system detection, and specifically relates to an underground space vehicle-mounted wall-penetrating radar path optimization method based on an improved ant colony algorithm. Background Art

[0002] Wall-penetrating radar is a device that uses low-frequency electromagnetic wave signals to detect targets behind walls or other shelters. It can penetrate and sense the information of the sheltered space behind the wall without being blocked by obstacles. It uses the strong penetration of ultra-wideband radar signals on non-metallic medium walls and other obstacles. After penetrating non-metallic obstacles such as walls, it generates radar echoes when encountering human targets. Weak movements such as human movement, breathing, heartbeat, and limb swings will form a Doppler effect on the echoes. Based on this, the received echoes are analyzed and processed to obtain information such as the location and number of targets.

[0003] With the acceleration of urbanization and the development and utilization of underground space, the demand for underground detection is increasing. At present, classic path planning algorithms, such as the A* algorithm, the Dijkstra algorithm, and the path search algorithm based on the Rapidly-exploring Random Tree (RRT), can have basic path planning capabilities in relatively simple environments, but in the highly uncertain and complex environment of underground space, they will face problems such as high computational complexity, easy to fall into local optimality, and low imaging quality; the traditional ant colony algorithm has strong global search capabilities and adaptability, but in practical applications, it still has problems such as slow convergence speed and insufficient path diversity. Therefore, in the complex and dynamically changing underground space, how to efficiently plan the motion path of the vehicle-mounted radar to improve the detection efficiency and imaging quality has become a key issue that needs to be solved urgently. Summary of the invention

[0004] To solve the above problems, the present invention proposes a path optimization method for underground space vehicle-mounted through-wall radar based on an improved ant colony algorithm. Aiming at the problems that the grating lobe effect exists in the three-dimensional tomographic through-wall radar imaging of traditional unmanned vehicles and the traditional algorithm cannot adapt to the complex environment of underground space, the present invention extends the principle of array optimization to suppress grating lobes to the unmanned vehicle-mounted through-wall radar, and realizes the joint optimization of the unmanned vehicle path length, non-uniform sampling and three-dimensional through-wall radar imaging quality based on the improved ant colony algorithm, so as to achieve high-quality imaging of targets behind the wall.

[0005] The specific technical solution of the present invention is as follows: A method for optimizing the path of an underground vehicle-mounted through-wall radar based on an improved ant colony algorithm, the steps of which are as follows:

[0006] Step 1: Perform linear interpolation processing on the known collision-free initial path to obtain a collision-free initial interpolation path, and then obtain path nodes with fixed intervals.

[0007] Step 2: Based on the demand for vehicle-mounted through-wall radar detection in underground spaces, the original cost function of the ant colony algorithm is improved, and an improved cost function is established by combining path length, imaging quality, and Gini impurity. The improved ant colony algorithm is then obtained to evaluate the quality of the path and obtain the optimized path.

[0008] Step 3: Use the path nodes with fixed intervals as the input of the improved ant colony algorithm, initialize the pheromone matrix, make the pheromone values ​​between each fixed interval path node equal, initialize the ant colony, and each ant represents a path for the unmanned vehicle to move.

[0009] Step 4: Use the improved ant colony algorithm to optimize the collision-free initial interpolation path, calculate the cost of each unmanned vehicle movement path, judge the quality of the path to update the pheromone, and enhance the probability of selecting the optimal path. Iteratively optimize the path node selection plan step by step until it converges to the optimal solution or reaches the preset number of iterations, and outputs the optimal path plan.

[0010] Step 5: Based on the optimal path solution, the vehicle-mounted wall-penetrating radar obtains the echo signal, and obtains the three-dimensional back-projection imaging result through pulse compression, distance interpolation, back-projection, and coherent accumulation processing.

[0011] Step 6: According to the three-dimensional back-projection imaging results, determine the wall position, thickness, and dielectric constant, and obtain the target position behind the wall.

[0012] Compared with the prior art, the present invention has the following significant advantages:

[0013] (1) The present invention can realize unmanned detection and target search and rescue in underground spaces.

[0014] (2) The present invention proposes an improved ant colony algorithm that takes into account the path length of the unmanned vehicle, the non-uniform sampling scheme and the quality of three-dimensional through-wall radar imaging, solves the joint optimization under multiple constraints, and achieves high-quality imaging of targets behind the wall.

[0015] (3) The present invention proposes an unmanned vehicle-mounted wall-penetrating detection method, which is based on the physical characteristics of electromagnetic waves, is not affected by narrow, dim environments, etc., has strong adaptability, and is an efficient wall-penetrating detection method.

[0016] (4) The present invention improves the path heuristic function, introduces an adaptive adjustment factor ξ, adaptively adjusts the weight of the heuristic information, balances the global search capability and convergence speed of the algorithm, and can obtain the optimal path for the unmanned vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a schematic diagram of the overall process of a method for optimizing the path of an underground vehicle-mounted through-wall radar based on an improved ant colony algorithm described in the present invention.

[0018] Figure 2 The figure is a flow chart of improving the ant colony algorithm of the present invention. DETAILED DESCRIPTION

[0019] The present invention is further described in detail below in conjunction with the accompanying drawings.

[0020] Combination Figure 1 The method for optimizing the path of an underground vehicle-mounted through-wall radar based on an improved ant colony algorithm of the present invention comprises the following steps:

[0021] Step 1: Perform linear interpolation processing on the known collision-free initial path to obtain a collision-free initial interpolation path, and then obtain path nodes with fixed intervals.

[0022] Considering the position of the unmanned vehicle in three-dimensional space, the known collision-free initial path is represented as P = {p 1 ,p 2 ,...,p i ,...,p n}, where pi = (xi, yi, zi) represents the three-dimensional space coordinates of the i-th collision-free initial path node, and n represents the total number of nodes.

[0023] Perform linear interpolation on the known collision-free initial path, determine the distance interval between the interpolation path nodes as Δd, and find the number of interpolation points k inserted between each pair of collision-free initial adjacent path points:

[0024]

[0025] Among them, p i+1 -p i represents the collision-free initial adjacent path point p i and p i+1 The Euclidean distance between them, floor() represents the rounding function.

[0026] The total number of interpolation path nodes is m=n+k·(n-1), and the number of nodes on the unmanned vehicle optimization path is set to M, M≤m.

[0027] For each pair of collision-free initial adjacent nodes p i and p i+1 The interpolation path node p obtained by inserting a nodes between ia The calculation formula is as follows:

[0028]

[0029] Among them, p i+1 Indicates that it is followed by p i The next node, p i+1=(x i+1 ,y i+1 ,z i+1 ), p ia represents the interpolation path node, p ia =(x ia ,y ia ,z ia ), a=1,2,...,k.

[0030] p ia =(x ia ,y ia ,z ia ) is calculated as:

[0031]

[0032]

[0033] Among them, x ia represents the interpolation path node p ia The x-axis coordinate of the a-th interpolation node in ia represents the interpolation path node p ia The y-axis coordinate of the a-th interpolation node in ia represents the interpolation path node p ia The z-axis coordinate of the a-th interpolation node in .

[0034] Repeat the above interpolation calculations to finally obtain a total path with a fixed interval after linear interpolation, that is, a collision-free initial interpolation path P':

[0035] P'={p 1 ,p 11 ,p 12 ,...,p 2 ,p 21 ,...,p nk ,p n}

[0036] Among them, P' contains the collision-free initial path node p 1 ,p 2 ,...,p n and the interpolation path node p 11 ,p 12 ,...p 1k ,p 21 ,...p 2k ,...p nk .

[0037] Step 2: Based on the demand for vehicle-mounted through-wall radar detection in underground spaces, the original cost function of the ant colony algorithm is improved, and an improved cost function is established by combining path length, imaging quality, and Gini impurity. The improved ant colony algorithm is then obtained to evaluate the quality of the path and obtain the optimized path.

[0038] Based on the demand for underground space vehicle-mounted through-wall radar detection, the cost function J is improved total for:

[0039] J total =w 1 ·D+w 2 ·R psl +w 3 ·Gini

[0040] Where D represents the optimized path length, R psl represents the peak-to-sidelobe ratio, Gini represents the Gini impurity, and w 1 To evaluate the hyperparameter weights of D, w 2 To evaluate R psl The hyperparameter weight, w 3 Hyperparameter weights for evaluating Gini.

[0041]

[0042] D is represented by the sum of the Euclidean distances between the nodes on the optimization path, and M is the number of nodes on the optimization path of the unmanned vehicle; (x h ,y h ,z h ) represents the three-dimensional space coordinates of the hth optimization path node, (x h+1 ,y h+1 ,z h+1 ) represents the three-dimensional space coordinates of the h+1th optimization path node.

[0043]

[0044] P sl The point set representing the grating lobe, P ml Represents the point set of the main lobe, R psl The smaller it is, the higher the image quality;

[0045] Gini Impurity

[0046] Gini is the randomness of unequally spaced sampling.

[0047] Where F represents the sum of the number of different motion path solutions for the unmanned vehicle, p f represents the proportion of samples belonging to scheme f, that is, where u fis the number of samples belonging to scheme f, that is, the number of sampling points in a specific interval; u is the total number of samples, that is, the number of all sampling points.

[0048] If the spacing is more evenly distributed, that is, most of the sampling point spacings fall in the same or adjacent intervals, the Gini impurity value will be low; if the sampling point spacings are more dispersed, that is, distributed in multiple different intervals, the Gini impurity value will be high.

[0049] High Gini impurity: indicates that the distribution of the sampling point spacing is relatively scattered, reflecting that the sampling points are highly random and the spacing distribution is not concentrated; this situation reflects that the sampling points are distributed in space in an uneven and relatively random manner, which is usually conducive to suppressing the grating lobe effect and improving the resolution of radar imaging.

[0050] Step 3: Use the path nodes with fixed intervals as the input of the improved ant colony algorithm, initialize the pheromone matrix, and make the pheromone values ​​between each fixed interval path node equal. Initialize the ant colony, and each ant represents a path for the unmanned vehicle to move. The details are as follows:

[0051] Each unmanned vehicle optimization path contains M path nodes, M≤m; each ant selects a path according to probability, which is determined by the pheromone concentration and the distance between the path nodes; the selection probability The calculation formula is:

[0052]

[0053] in, is the probability of the ant choosing from node b to node c at time t, τ bc (t) is the pheromone concentration of the ant from node b to node c at time t, η bc (t) is the path heuristic information of the ant at time t, α and β are adjustment parameters, s represents the node that has not been reached, s∈allowed m ; τ bs (t) is the pheromone concentration of the ant from node b to the unreached node s at time t, η bs (t) is the reciprocal of the distance between the ant from node b to the unreached node s at time t; allowed m Represents the set of nodes of the unmanned vehicle path.

[0054] η bc (t) is expressed as follows:

[0055]

[0056] Where ξ is the adaptive adjustment factor, σ 1 YesR psl The weight coefficient, σ 2 Yesbc The weight coefficient; λ is the attenuation coefficient, d bc is the Euclidean distance from node b to node c, iter is the number of iterations at time t, iter max is the maximum number of iterations, Is a constant value.

[0057] Step 4: Combine Figure 2 , the improved ant colony algorithm is used to optimize the collision-free initial interpolation path, calculate the cost of each unmanned vehicle movement path, judge the quality of the path to update the pheromone, and enhance the probability of selecting the optimal path. The selection scheme of the path nodes is gradually optimized through multiple iterations until it converges to the optimal solution or reaches the preset number of iterations, and the optimal path scheme is output, as follows:

[0058] In traditional ant colony algorithms, pheromones are often updated based on the path length, while the present invention updates the pheromones of the ant path based on the weighted sum of the path length of the unmanned vehicle, the Gini impurity and the imaging quality of the through-wall radar. When the cost function of an ant path is smaller, the more pheromones it releases on the path, and the probability that the next generation of ants will take the path is greater, so that the feasible solution generated in the iterative process gradually converges to the direction of reducing the cost function. The calculation formula for pheromone update is as follows:

[0059]

[0060] Among them, τ bc (t+1) represents the pheromone concentration between nodes at time t+1, τ bc (t) represents the pheromone concentration between nodes at time t, ρ is the pheromone volatility factor, and ρ∈(0,1), the size of ρ directly affects the global search ability and convergence speed of the ant colony; Δτ bc (t) is the amount of new pheromone at time t, which is usually related to the quality of the path, N c is the total number of ants; is the pheromone concentration between nodes of ant g at time t, and the pheromone increment Q 0 is a constant, the cost function of ant g D, R psl , the weighted sum of Gini, p b is the three-dimensional space coordinate of the bth collision-free initial path node, p c is the three-dimensional space coordinate of the cth collision-free initial path node.

[0061] Step 5: Based on the optimal path solution, the vehicle-mounted wall-penetrating radar obtains the echo signal, and obtains the three-dimensional back-projection imaging result through pulse compression, distance interpolation, back-projection, and coherent accumulation processing.

[0062] The processing process of radar echo signals is as follows: pulse compression is performed to improve the distance resolution; the distance error between sampling points is corrected by the distance interpolation algorithm; three-dimensional imaging is performed using the back-projection algorithm to reconstruct the spatial position of the target object; multiple echo signals are superimposed through the coherent accumulation technology to enhance the target signal and suppress noise.

[0063] The back-projection algorithm is as follows: Assuming that the airborne radar transmits a linear frequency modulated continuous wave, the single slow-time data imaging result will obtain a point spread function as shown in the following formula:

[0064]

[0065] Where Band is the radar bandwidth, C is the speed of light, μ is the radar wavelength, ΔR represents the difference between the “radar-target” spacing and the “radar-grating lobe” spacing, ΔR = P t -P a -(P s -P a ), P t is the target position, P a is the slow-time airborne radar position, P s is the position of the grating lobe, and j represents the imaginary part.

[0066] Step 6: According to the three-dimensional back-projection imaging results, determine the wall position, thickness, and dielectric constant, and obtain the target position behind the wall.

[0067] By measuring the time from the wall-penetrating radar transmitting to the receiving of the echo signal, combined with the known signal propagation speed, the wall position is calculated. The specific method is as follows:

[0068]

[0069] Among them, W d is the distance to the wall, C v is the propagation speed of electromagnetic waves in the air, and Δt is the round-trip time of the echo signal.

[0070] When the signal encounters a wall, part of the signal will be reflected, and part of the signal will penetrate the wall and reflect again. By analyzing the time delay of the echo signal, the thickness of the wall can be determined. The first reflection peak of the signal represents the front of the wall, and the second peak represents the back of the wall. The wall thickness can be calculated using the following formula:

[0071]

[0072] Among them, W h is the wall thickness, t 1 and t 2 are the arrival times of the two reflection peaks respectively.

[0073] The dielectric constant can be determined by analyzing the attenuation and phase change of the echo signal. According to the relationship between the dielectric constant and the speed of electromagnetic waves, the propagation speed of the known signal and the phase delay of the reflected signal can be calculated using the following formula:

[0074]

[0075] Among them, ε r is the dielectric constant of the wall, W v is the propagation speed of electromagnetic waves in the wall.

[0076] Through comprehensive analysis of the wall's position, thickness and dielectric constant, the target position behind the wall can be corrected according to these physical parameters.

[0077] The above contents are only for explaining the technical idea of ​​the present invention and cannot be used to limit the protection scope of the present invention. Any changes made to the technical solutions according to the technical idea proposed by the present invention shall fall within the protection scope of the claims of the present invention.

Claims

1. A path optimization method for underground vehicle-mounted through-wall radar based on an improved ant colony algorithm, characterized in that: The following steps are involved: Step 1: Perform linear interpolation processing on the known collision-free initial path to obtain a collision-free initial interpolation path, and then obtain path nodes with fixed intervals; Step 2: Based on the underground space vehicle-mounted through-wall radar detection requirements, the original cost function of the ant colony algorithm is improved, and the improved cost function is established by combining the path length, imaging quality, and Gini impurity, and then the improved ant colony algorithm is obtained to evaluate the quality of the path and obtain the optimized path; Step 3: Use the path nodes with fixed intervals as the input of the improved ant colony algorithm, initialize the pheromone matrix, make the pheromone values ​​between each fixed interval path node equal, initialize the ant colony, and each ant represents a path for the unmanned vehicle to move; Step 4: Use the improved ant colony algorithm to optimize the collision-free initial interpolation path, calculate the cost of each unmanned vehicle movement path, judge the quality of the path to update the pheromone, and enhance the probability of selecting the optimal path. Iterate multiple times to gradually optimize the selection plan of the path node until it converges to the optimal solution or reaches the preset number of iterations, and output the optimal path plan; Step 5: Based on the optimal path solution, the vehicle-mounted through-wall radar obtains the echo signal, and obtains the three-dimensional back-projection imaging result through pulse compression, distance interpolation, back-projection, and coherent accumulation processing; Step 6: According to the three-dimensional back-projection imaging results, determine the wall position, thickness, and dielectric constant, and obtain the target position behind the wall.

2. The underground space vehicle-mounted through-wall radar path optimization method based on improved ant colony algorithm according to claim 1 is characterized by: In step 1, linear interpolation is performed on the known collision-free initial path to obtain a collision-free initial interpolation path, and then path nodes with fixed intervals are obtained, as follows: Considering the position of the unmanned vehicle in three-dimensional space, the known collision-free initial path is represented as P = {p1, p2, ..., p i ,...,p n }, where p i =(x i ,y i ,z i ) represents the three-dimensional space coordinates of the i-th collision-free initial path node, and n represents the total number of nodes; Perform linear interpolation on the known collision-free initial path, determine the distance interval between the interpolation path nodes as Δd, and find the number of interpolation points k inserted between each pair of collision-free initial adjacent path points: Among them, ||p i+1 -p i || indicates collision-free initial adjacent path point p i and p i+1 The Euclidean distance between them, floor() represents the rounding function; The total number of interpolation path nodes m = n + k (n-1), set the number of nodes on the unmanned vehicle optimization path to M, M ≤ m; For each pair of collision-free initial adjacent nodes p i and p i+1 The interpolation path node p obtained by inserting a nodes between ia The calculation formula is as follows: Among them, p i+1 Indicates that it is followed by p i The next node, p i+1 =(x i+1 ,y i+1 ,z i+1 ), p ia represents the interpolation path node, p ia =(x ia ,y ia ,z ia ), a=1,2,...,k; p ia =(x ia ,y ia ,z ia ) is calculated as: Among them, x ia represents the interpolation path node p ia The x-axis coordinate of the a-th interpolation node in ia represents the interpolation path node p ia The y-axis coordinate of the a-th interpolation node in ia represents the interpolation path node p ia The z-axis coordinate of the a-th interpolation node in ; Repeat the above interpolation calculations to finally obtain a total path with a fixed interval after linear interpolation, that is, a collision-free initial interpolation path P': P'={p1,p 11 ,p 12 ,...,p2,p 21 ,...,p nk ,p n } Among them, P' contains the collision-free initial path nodes p1, p2, ..., p n and the interpolation path node p 11 ,p 12 ,...p 1k ,p 21 ,...p 2k ,...p nk .

3. The underground space vehicle-mounted through-wall radar path optimization method based on improved ant colony algorithm according to claim 2 is characterized in that: In step 2, based on the underground space vehicle-mounted through-wall radar detection requirements, the original cost function of the ant colony algorithm is improved, and the improved cost function is established by combining the path length, imaging quality, and Gini impurity. Then, the improved ant colony algorithm is obtained to evaluate the quality of the path and obtain the optimized path, as follows: Based on the demand for underground space vehicle-mounted through-wall radar detection, the cost function J is improved total for: J total =w1·D+w2·R psl +w3·This Where D represents the optimized path length, R psl represents the peak-to-sidelobe ratio, Gini represents the Gini impurity, w1 is the hyperparameter weight for evaluating D, and w2 is the hyperparameter weight for evaluating R psl The hyperparameter weight of , w3 is the hyperparameter weight for evaluating Gini; D is represented by the sum of the Euclidean distances between the nodes on the optimization path, and M is the number of nodes on the optimization path of the unmanned vehicle; (x h ,y h ,z h ) represents the three-dimensional space coordinates of the hth optimization path node, (x h+1 ,y h+1 ,z h+1 ) represents the three-dimensional space coordinates of the h+1th optimization path node; P sl The point set representing the grating lobe, P ml Represents the point set of the main lobe, R psl The smaller it is, the higher the image quality; Gini Impurity Gini is the randomness of unequally spaced sampling; Where F represents the sum of the number of different motion path solutions for the unmanned vehicle, p f represents the proportion of samples belonging to scheme f, that is, where u f is the number of samples belonging to scheme f, that is, the number of sampling points in a specific interval; u is the total number of samples, that is, the number of all sampling points.

4. The underground space vehicle-mounted through-wall radar path optimization method based on improved ant colony algorithm according to claim 3 is characterized by: In step 3, the path nodes with fixed intervals are used as the input of the improved ant colony algorithm to initialize the pheromone matrix. The pheromone values ​​between each fixed interval path node are equal. The ant colony is initialized, and each ant represents a path for the unmanned vehicle to move. The details are as follows: There are M path nodes on each unmanned vehicle optimization path, M≤m; each ant chooses a path according to probability, which is determined by the pheromone concentration and the distance between the path nodes; the choice probability The calculation formula is: in, is the probability of the ant choosing from node b to node c at time t, τ bc (t) is the pheromone concentration of the ant from node b to node c at time t, η bc (t) is the path heuristic information of the ant at time t, α and β are adjustment parameters, s represents the node that has not been reached, s∈allowed m ; τ bs (t) is the pheromone concentration of the ant from node b to the unreached node s at time t, η bs (t) is the reciprocal of the distance between the ant from node b to the unreached node s at time t; allowed m Represents the set of nodes for each unmanned vehicle path; η bc (t) is expressed as follows: Where ξ is the adaptive adjustment factor, σ1 is R psl The weight coefficient of σ2 is d bc The weight coefficient; λ is the attenuation coefficient, d bc is the Euclidean distance from node b to node c, iter is the number of iterations at time t, iter max is the maximum number of iterations, and l is a constant value.

5. The underground space vehicle-mounted through-wall radar path optimization method based on improved ant colony algorithm according to claim 4 is characterized by: In step 4, the improved ant colony algorithm is used to optimize the collision-free initial interpolation path and calculate the cost of each unmanned vehicle motion path, as follows: The problem of calculating the cost of each unmanned vehicle's motion path can be described as an optimization problem with multiple constraints: min J total s.t.R ps1 =B(D,Gini) Among them, B represents the imaging operator based on the current unmanned vehicle-mounted wall-penetrating radar motion path length D and Gini. The peak sidelobe ratio R of the image is calculated based on the unmanned vehicle motion path length D and Gini. ps1 , and obtain the optimal driving path under the current evaluation criteria by minimizing the joint objective function.

6. The underground space vehicle-mounted through-wall radar path optimization method based on improved ant colony algorithm according to claim 5 is characterized by: In step 4, the pheromone calculation is as follows: The formula for pheromone concentration is: t bc (t+1)=(1-ρ)·τ bc (t)+Δτ bc (t) Among them, τ bc (t+1) represents the pheromone concentration between nodes at time t+1; ρ is the pheromone volatility factor, and ρ∈(0,1), the size of ρ directly affects the global search ability and convergence speed of the ant colony; Δτ bc (t) is the amount of newly added pheromone at time t, which is usually related to the quality of the path.

7. The underground space vehicle-mounted through-wall radar path optimization method based on improved ant colony algorithm according to claim 6 is characterized in that: In step 5, the processing process of the radar echo signal by pulse compression, distance interpolation, back projection, and coherent accumulation is performed to obtain the three-dimensional back projection imaging result, which is specifically as follows: Based on the optimal path solution, the vehicle-mounted through-wall radar obtains the echo signal and performs pulse compression to improve the distance resolution; the distance error between sampling points is corrected by the distance interpolation algorithm; the back-projection algorithm is used for three-dimensional imaging to reconstruct the spatial position of the target object; and the coherent accumulation technology is used to superimpose multiple echo signals to enhance the target signal and suppress noise.

8. The underground space vehicle-mounted through-wall radar path optimization method based on improved ant colony algorithm according to claim 7 is characterized in that: Use the back-projection algorithm to perform three-dimensional imaging and reconstruct the spatial position of the target object, as follows: Assuming that the airborne radar transmits a linear frequency modulated continuous wave, the single slow-time data imaging result will obtain a point spread function PSF, as shown in the following formula: Where, Band is the radar bandwidth, C is the speed of light, μ is the radar wavelength, j represents the imaginary part, and ΔR represents the difference between the "radar-target" spacing and the "radar-grating lobe" spacing; ΔR=P t -P a -(P s -P a ) Among them, P t is the target position, P a is the slow-time airborne radar position, P s is the position of the grating lobe.

9. The underground space vehicle-mounted through-wall radar path optimization method based on improved ant colony algorithm according to claim 8 is characterized in that: In step 6, according to the three-dimensional back-projection imaging results, the wall position, thickness, and dielectric constant are determined, and the target position behind the wall is obtained, as follows: The dielectric constant is determined by the attenuation and phase delay of the radar echo signal, and the position and thickness of the wall are calculated by the reflection characteristics of the radar echo signal, thereby obtaining the corrected point target position behind the wall.

Images