A navigation and positioning method for unmanned transport platform in chicken houses based on under-cage feature extraction

Abstract

The present invention relates to the field of navigation and positioning technology, and more specifically to a method for navigating and positioning an unmanned transport platform in a chicken coop based on under-cage feature extraction. The method comprises the following steps: scanning the edge of the feeding trough under the cage using a linear array camera with adjustable pitch angles, extracting continuous geometric features based on Fourier descriptors, and generating a characteristic baseline consistent with the extension direction of the cage; kinematically binding the real-time motion parameters of the transport platform to the characteristic baseline, and dynamically generating a navigation path curvature constraint based on the curvature change of the characteristic baseline; and calculating the platform posture offset based on the feature matching degree at the trough joints between adjacent cages and inversely correcting the motion control parameters to achieve closed-loop posture calibration. The present invention achieves dynamic matching between path curvature and platform steering capability, improving the physical feasibility and stability of path tracking and adapting to the operational requirements of narrow passages and large curvature turns in complex chicken coop scenarios.

Description

Technical Field

[0001] The present invention relates to the field of navigation and positioning technology, and in particular to a navigation and positioning method for an unmanned transport platform in a chicken house based on under-cage feature extraction. Background Art

[0002] With the intensification and automation of the livestock farming industry, the feeding, inspection, and manure cleaning processes within chicken houses are increasingly dependent on the participation of unmanned transport platforms. However, the internal structures of chicken houses are generally dense and repetitive, with narrow spaces and complex lighting. Traditional navigation methods face many technical challenges:

[0003] On the one hand, navigation methods based on global positioning systems (such as GPS and UWB) are susceptible to interference from factors such as metal obstructions and closed spaces within the chicken house environment, leading to signal distortion or failure, making it difficult to meet the requirements of high-precision continuous navigation. On the other hand, navigation methods that rely on ground path markers, QR codes, magnetic strips, or artificial anchor points are prone to wear and tear, have high maintenance costs, and are difficult to adapt to dynamic environmental changes, making them unfeasible for long-term deployment.

[0004] In addition, although some current visual navigation methods can use environmental features for positioning, due to ground pollution, uneven lighting, and monotonous structure inside the chicken house, feature point extraction is unstable and positioning accuracy fluctuates greatly. Especially during the long-term operation of the platform, the accumulated trajectory error can easily lead to posture drift, and there is a lack of an effective closed-loop correction mechanism. Summary of the Invention

[0005] The present invention provides a navigation and positioning method for an unmanned transport platform in a chicken house based on under-cage feature extraction. The method can adapt to the structural layout of the chicken house, has stable feature extraction capabilities and highly robust path constraint control capabilities, and supports a navigation and positioning method with self-posture correction under continuous operation, so as to improve the operating efficiency and operational stability of the platform in complex breeding environments.

[0006] A method for navigating and positioning an unmanned transport platform in a chicken house based on under-cage feature extraction comprises the following steps:

[0007] S1: Scan the edge of the feeding trough under the cage with a linear array camera with adjustable pitch angle, extract continuous geometric features based on Fourier descriptors, and generate a feature baseline consistent with the extension direction of the cage;

[0008] S2: kinematically binding the real-time motion parameters of the carrier platform to the characteristic baseline, and dynamically generating a navigation path curvature constraint according to the curvature change of the characteristic baseline;

[0009] S3: Based on the feature matching degree of the feeding trough joints between adjacent cages, the platform posture offset is calculated and the motion control parameters are corrected inversely to achieve closed-loop posture calibration.

[0010] Optionally, the S1 specifically includes:

[0011] S11, adjusting the pitch angle of the linear array camera according to the cage height and the inclination angle of the trough installation so that the scanning plane is orthogonal to the extension direction of the trough edge;

[0012] S12, extracting the edge pixel point set of the trough in the contour sequence image, using Fourier descriptor to perform frequency domain encoding on the edge contour, retaining the first N order low frequency coefficients to reconstruct the geometric curve, specifically including extracting the edge pixel point set from each frame of the contour sequence image Perform frequency domain coding using the complex discrete Fourier transform: , Indicates the The horizontal coordinates of the edge pixels, Indicates the The vertical coordinates of edge pixels, is the imaginary unit, is the number of edge points, For the Fourier coefficients, retaining the previous The frequency order is used to reconstruct the smooth edge curve;

[0013] S13, based on the mechanical installation constraint relationship between adjacent cages, perform a spatial consistency check on the reconstructed geometric curve and select a continuous segment that meets the following conditions as a characteristic baseline:

[0014] The spacing error between adjacent cage base lines satisfies: ;

[0015] Curvature change rate Less than the set curvature threshold: ;

[0016] in, Allow tolerance for cage installation, is the vertical distance deviation between adjacent reference lines, is the local curvature of the curve, is the arc length coordinate parameter of the curve, is the curvature change rate;

[0017] Optionally, the pitch angle of the line array camera is calculated as: ;in, is the cage height, is the horizontal projection length of the trough, Install an inclination angle for the trough, The pitch angle that needs to be adjusted for the line scan camera.

[0018] Optionally, the S1 further includes continuously collecting contour sequence images of the edge of the trough along the traveling direction of the carrying platform through a linear array camera, and performing a morphological closing operation on each frame of the contour sequence image to eliminate interference from feed residues.

[0019] Optionally, the S2 specifically includes:

[0020] S21, real-time acquisition of platform linear speed , actual steering angle and heading angle , establishing an Ackerman steering kinematic model of the carrier platform including linear velocity, actual steering angle and heading angle, and obtaining motion parameters in real time through the carrier platform, wherein the Ackerman steering kinematic model describes the geometric mapping relationship between the platform steering behavior and the front wheel angle;

[0021] S22, extract the obtained characteristic baseline, perform local geometric fitting on the characteristic baseline at the current position of the carrier platform, obtain the corresponding expected path curvature, and convert the real-time curvature of the characteristic baseline into As the desired path curvature input, the desired path curvature is converted into the desired steering angle according to the kinematic model ;

[0022] S23, according to the actual steering angle and The difference , adjust the curvature constraint range of the navigation path ;

[0023] S24, based on constraint range Generate an allowable path curvature corridor and modify the corridor boundary in combination with the platform's maximum centripetal acceleration limit.

[0024] Optionally, the S23 specifically includes: based on , construct the curvature constraint range of the navigation path currently acceptable to the carrier platform, and the upper and lower boundaries of the constraint range are adjusted according to the steering response coefficient to reflect the actual response capability of the carrier platform under different speeds and adhesion conditions.

[0025] Optionally, in S22, the desired steering angle The calculation is expressed as:

[0026] ;in, is the desired steering angle, is the platform wheelbase, is the current curvature of the feature baseline, which is derived as:

[0027] .

[0028] Optionally, the S3 specifically includes:

[0029] S31, when the carrier platform moves to the joint of adjacent cage bodies, extract the edge feature point set of the joint area and calculate the Hausdorff distance matching degree between it and the preset standard joint template ;

[0030] S32, according to the matching degree Deviation relationship with the benchmark calibration threshold, solving the platform posture offset, including lateral offset and heading angle offset ;

[0031] S33, will and Input PID controller to generate steering angular velocity compensation coefficient and heading constraint correction ;

[0032] S34, reverse correction of actual steering angle control parameters and heading angle constraint parameters , completing the closed-loop pose calibration.

[0033] Optionally, in said S32, the platform posture offset is calculated based on the matching degree With the reference calibration threshold The deviation realization is expressed as:

[0034] ;

[0035] ;

[0036] in, is the lateral position offset of the platform, is the platform heading angle offset, is the image error-physical displacement conversion coefficient, is the heading deviation correction coefficient, is the longitudinal physical span of the joint region feature, is the threshold for seam matching.

[0037] Beneficial effects of the present invention:

[0038] The present invention uses a linear array camera with adjustable pitch angle to scan the edge of the feeding trough under the cage, and combines the Fourier descriptor to reconstruct the edge curve in the frequency domain. While eliminating feed interference, it retains a stable low-frequency contour morphology and selects smooth segments with a curvature change rate less than a threshold as a characteristic baseline. Compared with traditional navigation methods based on global structure or single-point marking, this method can continuously and stably extract geometric navigation references in closed environments with significant visual interference and high structural repeatability, thereby improving the continuity and anti-interference ability of platform positioning.

[0039] The present invention establishes a kinematic binding relationship of the Ackerman steering platform based on the real-time curvature of the characteristic baseline, adaptively generates a navigation curvature corridor through the difference between the theoretical steering angle and the actual steering response, and integrates the platform's maximum centripetal acceleration constraint to perform physical boundary correction, effectively avoiding the problem of path planning being out of touch with platform dynamics, achieving dynamic matching between path curvature and platform steering capability, improving the physical feasibility and stability of path tracking, and adapting to the operational requirements of narrow channels and large curvature turns in complex chicken house scenarios.

[0040] During platform operation, the present invention utilizes the trough seams between adjacent cages as a natural positioning reference, introduces the Hausdorff distance for template matching of seam edges, and calculates the platform's lateral and directional offsets based on the error magnitude, forming an offset resolution mechanism based on image-to-physical quantity conversion. Combined with the compensation generated by a PID controller, the original navigation control parameters are reversely corrected to achieve closed-loop self-calibration of the position and posture. This overcomes the trajectory offset issues caused by visual drift or error accumulation in traditional navigation, improving the long-term navigation accuracy and robustness in large-scale chicken coop environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0042] Figure 1 Schematic diagram of a method flow in an embodiment of the present invention;

[0043] Figure 2 Schematic diagram of feature baseline extraction according to an embodiment of the present invention. DETAILED DESCRIPTION

[0044] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments. It is also noted that, to provide a more detailed description, the following embodiments are best and preferred embodiments, and those skilled in the art may employ alternative methods for implementing certain known technologies. Furthermore, the accompanying drawings are intended only to provide a more detailed description of the embodiments and are not intended to limit the present invention.

[0045] It should be noted that references in the specification to "one embodiment," "an embodiment," "exemplary embodiments," "some embodiments," etc. indicate that the described embodiments may include specific features, structures, or characteristics, but not necessarily every embodiment will include such specific features, structures, or characteristics. Furthermore, when specific features, structures, or characteristics are described in conjunction with an embodiment, it is within the knowledge of persons skilled in the relevant art to implement such features, structures, or characteristics in conjunction with other embodiments (whether or not explicitly described).

[0046] In general, terms can be understood, at least in part, from their use in context. For example, depending at least in part on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending at least in part on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0047] like Figure 1-Figure 2 As shown, a method for navigating and positioning an unmanned transport platform in a chicken house based on under-cage feature extraction includes the following steps:

[0048] S1: Scan the edge of the feeding trough under the cage with a linear array camera with adjustable pitch angle, extract continuous geometric features based on Fourier descriptors, and generate a feature baseline consistent with the extension direction of the cage;

[0049] S2: Kinematically bind the real-time motion parameters of the carrier platform to the characteristic baseline, and dynamically generate the curvature constraint of the navigation path according to the curvature change of the characteristic baseline;

[0050] S3: Based on the feature matching degree of the feeding trough joints between adjacent cages, the platform posture offset is calculated and the motion control parameters are corrected inversely to achieve closed-loop posture calibration.

[0051] S1 specifically includes:

[0052] S11, dynamic adjustment mechanism of pitch angle: according to the height of the cage And the trough installation angle ,Dynamically adjust the pitch angle of the linear array camera , so that the scanning plane is orthogonal to the extension direction of the trough edge, the pitch angle calculation formula is:

[0053] ;in, is the cage height, is the horizontal projection length of the trough, Install an inclination angle for the trough, The pitch angle that needs to be adjusted for the line scan camera.

[0054] S12, contour image anti-interference processing: The linear array camera continuously collects contour sequence images of the trough edge along the direction of travel of the carrier platform, and performs morphological closing operation on each frame of the image to eliminate interference from feed residue. The operating parameters include:

[0055] Closing operation structure element size ,in, , i.e. the maximum particle size of feed particles (measured value);

[0056] The scanning synchronization relationship satisfies: ,in, is the line frequency of the line scan camera, is the operating speed of the carrier platform, is the physical spatial resolution corresponding to the pixel.

[0057] S13, Fourier descriptor feature extraction and reconstruction: edge pixel points extracted from each frame image (in ) is used for frequency domain coding, using the discrete Fourier transform in complex form: , Indicates the The horizontal coordinates of the edge pixels, Indicates the The vertical coordinates of edge pixels, is an imaginary unit, which is used to express the two-dimensional coordinates as the pixel points at the edge of the trough in complex form so as to perform a discrete Fourier transform in complex form;

[0058] Reconstruct the curve to keep the previous The empirical value of the low-frequency coefficient is: (corresponding reconstruction error <0.5mm), where, is the number of edge points, For the Resampled complex edge points, For the The order Fourier coefficients, is the frequency order retained, used to reconstruct the smooth edge curve.

[0059] S14, spatial consistency check and baseline screening: Based on the mechanical installation constraints of the cage, the reconstructed geometric curve is checked for consistency, and continuous segments that meet the following conditions are selected as feature baselines:

[0060] The spacing error between adjacent cage base lines satisfies: ;

[0061] Curvature change rate Less than the set curvature threshold: ;

[0062] in, Allow tolerance for cage installation, is the vertical distance deviation between adjacent reference lines, is the local curvature of the curve, is the arc length coordinate parameter of the curve, is the curvature change rate (used to reflect the smoothness of the curve), and the threshold corresponds to the minimum turning radius .

[0063] S2 specifically includes:

[0064] S21, Kinematic Model Binding: Establish the Ackerman steering kinematic model of the carrier platform and obtain the linear velocity of the platform in real time , actual steering angle and heading angle , where data source: actual steering angle Provided by the vehicle's CAN bus.

[0065] The Ackermann steering kinematic model is based on the non-slip assumption and is applicable to front-wheel steering, rear-wheel drive, or four-wheel drive platforms. Its structure is defined as follows:

[0066] 1. The center of mass is used as the reference point to model the two-dimensional motion of the platform in the plane. The front wheels are assumed to rotate at the same angle. The rear wheels rotate parallel to the longitudinal axis of the platform, and the platform does not slip. The state variables defining the center of mass of the vehicle include: plane position (x, y), heading angle ψ, linear velocity v, actual steering angle And the wheelbase Lwheelbase.

[0067] 2. State equation (kinematic differential equation): The platform motion can be described by the following differential equation:

[0068] The change of center of mass position is determined by linear velocity and heading angle;

[0069] The heading angle change is determined by the steering angle;

[0070] The angular velocity of the platform is a combination of the linear velocity and the steering angle.

[0071] These equations are used for attitude prediction or path inversion in discrete control or simulation processes.

[0072] 3. Parameter acquisition:

[0073] Linear velocity v: measured in real time by the encoder (installed on the drive wheel shaft) and obtained by calculating the distance the wheel rotates per unit time.

[0074] Actual steering angle : Measured in real time by a steering angle sensor or angle potentiometer and read via the CAN bus or other communication interface;

[0075] Heading angle ψ: obtained by integrating the gyroscope in the inertial navigation system (IMU). A more stable heading estimate can also be obtained by fusing GPS, magnetometer, and odometer.

[0076] After the kinematic model is established, the actual 、 、 The value is compared with the reference value generated in path planning to perform trajectory tracking control.

[0077] S22, expected steering angle calculation: the real-time curvature of the characteristic baseline As the desired path curvature input, the expected steering angle of the platform under the path curvature is solved according to the Ackerman model. :

[0078] ;in, is the desired steering angle, is the platform wheelbase (m), is the current curvature of the feature baseline, which is derived as follows:

[0079] .

[0080] S23, dynamic curvature constraint range generation:

[0081] Calculate the difference between the actual steering angle and the desired steering angle: ;

[0082] Adjust the path curvature constraint range based on the difference: ;

[0083] in, is the platform steering response coefficient, is the permissible range of curvature of the navigation path, and the platform steering response coefficient It can be expressed as: ,in, is the platform linear velocity, is the acceleration due to gravity (9.81m / s 2 ), The friction coefficient between the road surface and the wheels dynamically adjusts the path curvature bandwidth to adapt to the response delay of the steering mechanism and the vehicle stability requirements.

[0084] S24, fusion of physical constraints to generate allowable curvature corridor: within the generated curvature constraint range Introduce the platform centripetal acceleration limit and correct the boundary: ,in, is the maximum path curvature that the platform can withstand, ensuring that the platform will not slip or become unstable due to excessive centrifugal force when following the curvature change. The final navigation path curvature corridor will be , used by the controller to dynamically generate feasible navigation paths.

[0085] The output of S1 is a geometric reference line segment that has been reconstructed by Fourier descriptor smoothing and screened by consistency check, that is, a two-dimensional continuous curve segment representing the edge contour of the trough in the image space, with small curvature fluctuations and a high-confidence navigation reference line representing the extension direction of the cage; the input of S2 is the curvature value corresponding to each local window or current frame in the reference line , as the expected path curvature.

[0086] The method for obtaining is as follows: The feature baseline is a curve composed of a series of smooth pixel points, through which the local curvature can be calculated:

[0087] ;in, Indicates the two-dimensional coordinate point on the baseline, according to the arc length parameterization, is the first-order derivative, representing the tangent direction, are second-order derivatives, representing the change in curvature direction, and these derivative values ​​are obtained by numerically differentiating the polynomial fit (cubic spline) to the baseline.

[0088] In S1 It is used to evaluate the curvature change of the entire curve at multiple consecutive points and to screen the sections with stable curvature. In the stable curve segment selected by S1, the curvature of the current position of the platform is extracted as the control input. Small, indicating that the curve The platform moves smoothly and continuously during the process, which is suitable for use as a navigation path. At this time, the path extracted from any point in the segment (such as the current position of the platform) The value is more representative, the error is smaller, and the control is more stable. Is satisfying The curvature value of the navigation input point selected under the threshold condition.

[0089] S3 specifically includes:

[0090] S31, feature seam matching calculation: When the carrier platform moves to the adjacent cage seam area, extract the edge feature point set of the area , and the preset standard seam template feature point set For comparison, Hausdorff distance is used for matching measurement:

[0091] ;

[0092] in, is the Hausdorff distance matching degree between the current seam area and the template (pixels), is the set of seam edge feature points extracted for the current frame, is the standard template feature point set, is the Euclidean distance.

[0093] S32, platform posture offset calculation: according to the matching degree The reference calibration threshold set by the system Deviation, solve the lateral offset of the platform and heading angle offset :

[0094] ;

[0095] ;

[0096] in, is the lateral position offset of the platform, is the platform heading angle offset, is the image error-physical displacement conversion coefficient, is the heading deviation correction coefficient. When the carrier platform has a normal speed of 0.4–0.6 m / s and an attitude control response time of < 0.5 s, the optimal correction result is , is the longitudinal physical span of the joint region feature, is the threshold for seam matching. It is used to convert the Hausdorff distance (image matching deviation in pixels) into the actual lateral physical offset of the platform. The acquisition method is through calibration experiments: multiple cage joint physical samples with known lateral offsets are arranged on both sides of the track ( ); Use the linear array camera installed on the platform to collect the image seam features of each sample in a fixed posture; calculate the Hausdorff distance for each group of samples ; Each known offset Corresponding Create a set of calibration data points; perform a univariate linear regression: The fitting slope is , residual Control within acceptable limits mm. In the actual measurement of the present invention, the imaging resolution of the line array camera is 0.25 mm / pixel. After calibration and fitting, we get: .

[0097] Benchmark calibration threshold The determination method is as follows: 100 sets of real seam images are collected in a controlled environment, including 80 sets of standard seam samples with intact structure and no offset, and 20 sets of non-standard samples with known offset or tilt errors. The Hausdorff distance between all samples and the standard template is calculated. , respectively obtain the maximum distance of the standard group and the minimum distance of the non-standard group , take the middle value between the two as the classification threshold: .

[0098] Experimental sample results (actually measured): Standard sample maximum distance: pixels; minimum distance for non-standard samples: pixels, so the matching threshold is: .

[0099] S33, generating steering and heading compensation parameters: and Input PID controller to generate steering angle compensation coefficient and heading angle correction :

[0100] ;

[0101] ;

[0102] in, is the multiplicative compensation factor of the steering angular velocity, is the additive correction of heading angle, are the proportional, integral and differential gain coefficients of the PID controller respectively, is the heading angle correction coefficient, is the original heading angle.

[0103] S34, navigation control parameter closed loop correction: Based on the above compensation amount, the actual steering angle control parameter and heading angle constraint parameters Make corrections and obtain the control parameters after closed-loop correction:

[0104] ;

[0105] ;

[0106] in, is the target steering angle after closed-loop correction, The above two quantities will be fed back to the controller to realize dynamic correction of path deviation.

[0107] The preset standard seam template feature point set is specifically constructed as follows:

[0108] 1. Acquisition method: The standard joint template feature point set is obtained by selecting multiple images of actual cage joint areas with complete structures and high installation accuracy under the working conditions after the cage mechanical installation is completed, and using Canny edge detection + morphological filtering to extract edge features.

[0109] 2. Point set structure: The standard template is a two-dimensional coordinate point set, denoted as:

[0110] ,in, For the number of feature points in the template, 50-100 representative edge points are selected. The feature points cover the typical geometric forms of seam segments in the image, such as vertical fractures and edge transition inflection points. All point sets are normalized and smoothed to improve matching stability and noise resistance.

[0111] 3. Feature alignment strategy: Use centroid alignment and principal direction alignment (PCA direction) to pre-register the template and the current image seam area to ensure that the Hausdorff distance reflects the actual geometric deviation.

[0112] The present invention encompasses any alternatives, modifications, equivalents, and solutions that fall within the spirit and scope of the present invention. To provide a thorough understanding of the present invention, specific details are described in detail below in connection with the preferred embodiments of the present invention, but those skilled in the art will be able to fully understand the present invention without these detailed descriptions. Furthermore, to avoid unnecessary confusion regarding the essence of the present invention, well-known methods, processes, procedures, components, and circuits have not been described in detail.

[0113] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A navigation and positioning method for an unmanned transport platform in a chicken house based on under-cage feature extraction, characterized in that: The following steps are involved: S1: Scan the edge of the feeding trough under the cage using a linear array camera with adjustable pitch angles. Continuous geometric features are extracted based on Fourier descriptors to generate a feature baseline consistent with the cage extension direction. Specifically, the following steps are performed: S11, adjusting the pitch angle of the linear array camera according to the cage height and the inclination angle of the trough installation so that the scanning plane is orthogonal to the extension direction of the trough edge; S12, extracting the edge pixel point set of the trough in the contour sequence image, using Fourier descriptor to perform frequency domain encoding on the edge contour, retaining the first N order low frequency coefficients to reconstruct the geometric curve, specifically including extracting the edge pixel point set from each frame of the contour sequence image Perform frequency domain coding using the complex discrete Fourier transform: , Indicates the The horizontal coordinates of the edge pixels, Indicates the The vertical coordinates of the edge pixels, is the imaginary unit, is the number of edge points, For the Fourier coefficients, retaining the previous The frequency order is used to reconstruct the smooth edge curve; S13, based on the mechanical installation constraint relationship between adjacent cages, perform a spatial consistency check on the reconstructed geometric curve and select a continuous segment that meets the following conditions as a characteristic baseline: The spacing error between adjacent cage base lines satisfies: ; Curvature change rate Less than the set curvature threshold: ; in, Allow tolerance for cage installation, is the vertical distance deviation between adjacent reference lines, is the local curvature of the curve, is the arc length coordinate parameter of the curve, is the curvature change rate; S2: kinematically binding the real-time motion parameters of the carrier platform to the characteristic baseline, and dynamically generating a navigation path curvature constraint according to the curvature change of the characteristic baseline; S3: Based on the feature matching degree of the feeding trough joints between adjacent cages, the platform posture offset is calculated and the motion control parameters are corrected inversely to achieve closed-loop posture calibration.

2. The method for navigating and positioning a chicken house unmanned transport platform based on under-cage feature extraction according to claim 1, characterized in that: The pitch angle of the linear array camera is calculated as: ;in, is the cage height, is the horizontal projection length of the trough, Install an inclination angle for the trough, The pitch angle that needs to be adjusted for the line scan camera.

3. The method for navigating and positioning a chicken house unmanned transport platform based on under-cage feature extraction according to claim 1, characterized in that: The S1 also includes continuously collecting contour sequence images of the edge of the trough along the traveling direction of the carrying platform through a linear array camera, and performing a morphological closing operation on each frame of the contour sequence image to eliminate interference from feed residues.

4. The method for navigating and positioning a chicken house unmanned transport platform based on under-cage feature extraction according to claim 1, characterized in that: The S2 specifically includes: S21, real-time acquisition of platform linear speed , actual steering angle and heading angle , establishing an Ackerman steering kinematic model of the carrier platform including linear velocity, actual steering angle and heading angle, and obtaining motion parameters in real time through the carrier platform, wherein the Ackerman steering kinematic model describes the geometric mapping relationship between the platform steering behavior and the front wheel angle; S22, extract the obtained characteristic baseline, perform local geometric fitting on the characteristic baseline at the current position of the carrier platform, obtain the corresponding expected path curvature, and convert the real-time curvature of the characteristic baseline into As the desired path curvature input, the desired path curvature is converted into the desired steering angle according to the kinematic model ; S23, according to the actual steering angle and The difference , adjust the curvature constraint range of the navigation path ; S24, based on constraint range Generate an allowable path curvature corridor and modify the corridor boundary in combination with the platform's maximum centripetal acceleration limit.

5. The method for navigating and positioning a chicken house unmanned transport platform based on under-cage feature extraction according to claim 4 is characterized in that: The S23 specifically includes: based on , construct the curvature constraint range of the navigation path currently acceptable to the carrier platform, and the upper and lower boundaries of the constraint range are adjusted according to the steering response coefficient to reflect the actual response capability of the carrier platform under different speeds and adhesion conditions.

6. The method for navigating and positioning a chicken house unmanned transport platform based on under-cage feature extraction according to claim 4, characterized in that: In the above S22, the desired steering angle The calculation is expressed as: ;in, is the desired steering angle, is the platform wheelbase, is the current curvature of the feature baseline, which is derived as: 。 7. The method for navigating and positioning a chicken house unmanned transport platform based on under-cage feature extraction according to claim 1, characterized in that: The S3 specifically includes: S31, when the carrier platform moves to the joint of adjacent cage bodies, extract the edge feature point set of the joint area and calculate the Hausdorff distance matching degree between it and the preset standard joint template ; S32, according to the matching degree Deviation relationship with the benchmark calibration threshold, solving the platform posture offset, including lateral offset and heading angle offset ; S33, will and Input PID controller to generate steering angular velocity compensation coefficient and heading constraint correction ; S34, reverse correction of actual steering angle control parameters and heading angle constraint parameters , completing the closed-loop pose calibration.

8. The method for navigating and positioning a chicken house unmanned transport platform based on under-cage feature extraction according to claim 7, characterized in that: In the S32, the platform posture offset is calculated based on the matching degree. With the reference calibration threshold The deviation realization is expressed as: ； ； in, is the lateral position offset of the platform, is the platform heading angle offset, is the image error-physical displacement conversion coefficient, is the heading deviation correction coefficient, is the longitudinal physical span of the joint region feature, is the threshold for seam matching.

Images