A path following control method applied to unmanned ground mobile platform

Abstract

The application discloses a path following control method applicable to an unmanned ground mobile platform, in which a traditional fixed preview distance scheme is adjusted into a real-time changing preview distance scheme, so as to adapt to the switching of straight running and turning, in a specific solving process, a lateral deviation and a longitudinal deviation are comprehensively considered, and weight values are respectively set, so as to improve the coincidence degree with a planned path, the method can greatly improve the precision of path following, realizes high-precision reproduction of an original given path, is favorable to improving the path reproduction precision and running reliability of the unmanned ground mobile platform, and has important practical significance and engineering application value.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of unmanned ground mobile platform control, in particular to a path following control method applied to an unmanned ground mobile platform. BACKGROUND

[0002] With the acceleration of the intelligent era, the application scenarios of unmanned ground mobile platforms are becoming more and more extensive, such as automatic transfer vehicles in large warehouses, etc. An unmanned ground mobile platform is generally provided with a path planning system, which can plan a reasonable walking route according to its own position and target position, i.e. a planned path. Path following is a major function of an unmanned ground mobile platform, i.e. walking along the planned path according to the planned path. For a general unmanned ground mobile platform, its path following system often has the following problems, i.e. the unmanned ground mobile platform has low accuracy in replicating the original path when following the path, and sometimes may even directly deviate from the original path. Deviating from the planned path will reduce transportation efficiency and may even cause traffic accidents such as scratching.

[0003] Due to the above reasons, the present inventors have made in-depth research on the path following control method of the existing unmanned ground mobile platform in the hope of designing a new control method that can solve the above problems. SUMMARY

[0004] In order to overcome the above problems, the present inventors have made intensive research and designed a path following control method applicable to an unmanned ground mobile platform. In this method, the traditional fixed preview distance scheme is adjusted to a real-time variable preview distance scheme in order to adapt to the switching between straight driving and turning. In the specific calculation process, the lateral deviation and longitudinal deviation are comprehensively considered and weight values are set respectively in order to improve the coincidence degree with the planned path. This method can greatly improve the accuracy of path following and realize high-precision replication of the original given path, which is beneficial to improving the path replication accuracy and driving reliability of the unmanned ground mobile platform and has important practical significance and engineering application value, thus completing the present application.

[0005] Specifically, the present application aims to provide a path following control method applied to an unmanned ground mobile platform, characterized in that,

[0006] The method comprises the following steps:

[0007] Step 1, obtaining the current preview distance in real time,

[0008] Step 2, obtaining the front wheel turning angle δ(t) according to the current preview distance value,

[0009] Step 3, outputting the δ(t) value to the front wheel steering actuator to control the front wheel steering accordingly.

[0010] wherein, step 1, step 2 and step 3 are performed once in each sampling period.

[0011] wherein, step 1 comprises the following sub-steps:

[0012] Sub-step 1, traverse the value of the preview distance l and substitute it into formula (I), each l value corresponds to a comprehensive deviation e(i+1) at the i+1 moment; d d

[0013] Sub-step 2, select the preview distance l that makes the absolute value of the comprehensive deviation e(i+1) at the i+1 moment minimum as the current preview distance; d

[0014] The formula (I) is:

[0015] e(i+1) = K1(i+1) + f(v) + f(v, l d ) (I)

[0016] wherein, e(i+1) represents the comprehensive deviation at the i+1 moment;

[0017] Preferably, K1(i+1) is obtained by the following formula (II), f(v) is obtained by the following formula (III), and f(v, l d ) is obtained by the following formula (IV);

[0018] K1(i+1) = w1(g e (i) - N(i+1)) + w2(g n (i) - E(i+1)) (II)

[0019] f(v) = w1vt0 cos θ(i) + w2vt0 sin θ(i) (III)

[0020]

[0021] wherein, w1 represents the weight of the lateral deviation, and w2 represents the weight of the longitudinal deviation;

[0022] (g e (i), g n (i)) represents the position coordinates of the rear wheel at the i moment;

[0023] (E(i+1), N(i+1)) represents the point on the planned path closest to the position coordinates (g e (i+1), g n (i+1)) of the rear wheel at the i+1 moment;

[0024] v represents the speed of the unmanned ground mobile platform;​​​

[0025] t0 represents a sampling period;

[0026] θ(i) represents a heading angle at the i-th moment;

[0027] (E aim , N aim ) represents a preview point coordinate at the i-th moment;

[0028] l d represents a preview distance.

[0029] wherein, in sub-step 1, the preview distance l d is in the range of 0.5m≤l d ≤5m; the value of l d has a precision of 0.1m in the traversal value process.

[0030] wherein, the weights w1 and w2 are obtained in real time through the following steps:

[0031] Step a, detecting a point on the given planning trajectory in a circular region with the rear wheel position (g e (i), g n (i)) of the ground mobile platform as the center and the minimum preview distance l dmin as the radius;

[0032] Step b, if there is no point on the planning trajectory in the circular region, then w1=w2=0.5;

[0033] if there is a point on the planning trajectory in the circular region, then find a point (E1, N1) with the minimum lateral deviation e1 from the center position and a point (E2, N2) with the minimum longitudinal deviation e2 from the center position from the points on the planning trajectory in the circular region, and calculate the minimum lateral deviation e 1min and the minimum longitudinal deviation e 2min ;

[0034] when e 1min ≥e 2min , w1=0.9, w2=0.1

[0035] when e 1min <e 2min , w1=0.8, w2=0.2.

[0036] wherein, a sensor is arranged on the unmanned ground mobile platform to detect the speed, rear wheel position coordinate and heading angle of the unmanned ground mobile platform in real time.

[0037] wherein, the (E aim , N aim) is obtained by the following formula (five):

[0038] (g e (i)-E aim ) 2 +(g n (i)-N aim ) 2 =l d 2 (five).

[0039] Wherein, in step 2, the front wheel steering angle δ(t) is obtained by the following formula (six):

[0040]

[0041] Wherein, δ(t) represents the front wheel steering angle, L represents the wheelbase of the ground mobile platform double axle, (g e , g n ) represents the rear wheel position coordinates, (g x , g y ) represents the preview point coordinates, θ(t) represents the heading angle, l d represents the preview distance.

[0042] Wherein, in step 3, it is judged whether the front wheel steering angle δ(t) obtained in step 2 is greater than the steering extreme value,

[0043] When the current front wheel steering angle δ(t) is greater than the steering extreme value, the steering extreme value is input into the execution mechanism of the front wheel steering,

[0044] When the current front wheel steering angle δ(t) is less than or equal to the steering extreme value, the current front wheel steering angle δ(t) is input into the execution mechanism of the front wheel steering.

[0045] The present application has the beneficial effects including:

[0046] (1) According to the path following control method applicable to the unmanned ground mobile platform provided by the present application, the comprehensive deviation control including lateral deviation and longitudinal deviation, makes the algorithm more comprehensive in theory;

[0047] (2) According to the path following control method applicable to the unmanned ground mobile platform provided by the present application, the weight allocation scheme of lateral deviation and longitudinal deviation is set, which can fully adapt to the tracking analysis of different road conditions, improve the following efficiency, and improve the coincidence degree with the planned path;

[0048] (3) According to the path following control method applicable to the unmanned ground mobile platform provided by the present application, by setting the preview distance as a dynamic value that changes in real time, the robustness of the system can be improved, the following effect can be improved, and the planned path can be approached to the greatest extent on both curved path and straight path. BRIEF DESCRIPTION OF DRAWINGS

[0049] Figure 1 This paper shows a comparison between the following path obtained when using a fixed pre-aiming distance in Embodiment 1 of this application and the original given path;

[0050] Figure 2 Show Figure 1 A graph showing the lateral deviation variation along the path.

[0051] Figure 3 This paper shows a comparison between the following path obtained in Embodiment 1 of this application, which takes into account lateral and longitudinal deviations and calculates the aiming distance in real time, and the original given path;

[0052] Figure 4 Show Figure 3 A graph showing the lateral deviation variation along the path.

[0053] Figure 5 This paper shows a comparison between the following path obtained in Embodiment 1 of this application, which takes into account lateral deviation and calculates the aiming distance in real time, and the original given path;

[0054] Figure 6 Show Figure 5 A graph showing the lateral deviation variation along the path.

[0055] Figure 7 This paper shows a comparison between the following path obtained when using a fixed pre-aiming distance in Embodiment 2 of this application and the original given path;

[0056] Figure 8 Show Figure 7 A graph showing the lateral deviation variation along the path.

[0057] Figure 9 This paper shows a comparison between the following path obtained in Embodiment 2 of this application, which takes into account lateral and longitudinal deviations and calculates the aiming distance in real time, and the original given path;

[0058] Figure 10 Show Figure 9 A graph showing the lateral deviation variation along the path.

[0059] Figure 11 This paper shows a comparison between the following path obtained in Embodiment 2 of this application, which takes into account lateral deviation and calculates the aiming distance in real time, and the original given path;

[0060] Figure 12 Show Figure 11 A graph showing the lateral deviation of the path being followed. Detailed Implementation

[0061] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Through these descriptions, the features and advantages of the present invention will become clearer and more apparent.

[0062] The term "exemplary" is used herein to mean "serving as an example, instance, or illustration." Any implementation described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other implementations. Unless specifically indicated otherwise, the drawings are not necessarily to scale.

[0063] In the present application, the unmanned ground mobile platform is provided with a path planning system and a following control system, wherein the path planning system plans a planning path according to a target position, aiming at making the mobile platform reach the target position along the planning path, and the specific action process is controlled by the following control system. In the general following control system, the speed of the mobile platform does not need to be controlled when no obstacle is encountered, and the mobile platform travels at a constant speed. The following control system only needs to control the steering of the mobile platform in real time, that is, the front wheel steering angle of the mobile platform is controlled in real time.

[0064] The unmanned ground mobile platform in the present application belongs to a low-speed mobile platform, and the moving speed thereof is generally below 10 km / h.

[0065] The present application provides a path following control method applied to an unmanned ground mobile platform, and the method comprises the following steps.

[0066] Step 1: obtaining a current preview distance in real time,

[0067] In a preferred embodiment, step 1 comprises the following substeps.

[0068] Substep 1: traversing the value of the preview distance l d and substituting it into formula (I), so that each l d value corresponds to a comprehensive deviation e(i+1) at the i+1 moment; the traversal

[0069] Substep 2: selecting the preview distance l d that makes the absolute value of the comprehensive deviation e(i+1) at the i+1 moment minimum as the current preview distance;

[0070] The formula (I) is as follows:

[0071] e(i+1)=K1(i+1)+f(v)+f(v,l d ) (I)

[0072] Wherein, e(i+1) represents the comprehensive deviation at the i+1 moment.

[0073] Preferably, K1(i+1) is obtained by formula (II) below, f(v) is obtained by formula (III) below, and f(v, l d ) is obtained by formula (IV) below.

[0074] K1(i+1) = w1(g e (i)-N(i+1))+w2(g n (i)-E(i+1)) (two)

[0075] f(v) = w1vt0 cos θ(i) + w2vt0 sin θ(i) (three)

[0076]

[0077] In the present application, formula (two) is a time-dependent quantity, formula (three) is a function of the ground mobile platform running speed v, and formula (four) is a function of the ground mobile platform running speed v and the preview distance l d , so the comprehensive deviation can be obtained by giving the specific parameters in the formula and the traversal range of the preview distance.

[0078] wherein w1 represents the weight of the lateral deviation, and w2 represents the weight of the longitudinal deviation;

[0079] (g e (i), g n (i)) represents the position coordinates of the rear wheel at the i-th moment;

[0080] When the unmanned ground mobile platform is provided with 4 or more wheels, the position coordinates of the rear wheel in the present application refer to the center position of the connecting shaft between the last two wheels.

[0081] The coordinates in the present application all refer to the coordinates in the ground coordinate system, and further, since the unmanned ground mobile platform moves on the ground, all the coordinates in the present application are coordinates on the ground plane, and the coordinate axes thereof are the X axis and the Y axis.

[0082] (E(i+1), N(i+1)) represents the point closest to the position coordinates (g e (i+1), g n (i+1)) of the rear wheel at the i+1-th moment on the planned path; the planned path used in the present application includes a plurality of spaced coordinate points, and the distance between two adjacent coordinate points is 0.5-1 meter, preferably 0.6 meter;

[0083] v represents the speed of the unmanned ground mobile platform; the speed is low, and the size thereof is generally below 10 km / h;

[0084] t0 represents the sampling period, and the value thereof is 0.05 s;

[0085] θ(i) represents the heading angle at the i-th moment;

[0086] In the present application, sampling and solution control are performed once at each time, i.e. the interval between the ith time and the (i+1)th time is 0.05s;

[0087] (E aim , N aim ) represents the preview point coordinates at the ith time;

[0088] l d represents the preview distance.

[0089] The preview point coordinates described in the present application refer to the point coordinates matched to the rear wheel position coordinates along the advancing direction of the unmanned ground mobile platform in the given original planning path, and the straight line distance between the unmanned ground mobile platform rear wheel position and the preview point is equal to the preview distance.

[0090] The preview distance described in the present application is also called the forward-looking distance, which refers to the straight line distance between the rear wheel position of the unmanned ground mobile platform and the preview point.

[0091] In a preferred embodiment, in sub-step 1, the preview distance l d is set to a value in the range of 0.5m≤l d ≤5m; the value precision of l d is 0.1m during the traversal value process. That is, in sub-step 1, the value of l d is set one by one, for example, 0.5m, 0.6m, 0.7m, …, 4.9m, 5m. The present inventors have found through repeated research that the traversal range of the preview distance is related to the size of the unmanned ground mobile platform itself, and the range defined in the present application can adapt to common size ground mobile platforms, the width of which is allowed to be between 0.5-1.5m, and the length of which can be between 1-2.5m. The precision of the preview distance in the traversal process is 0.1m, which can achieve accurate following. If it is adjusted to 0.01m, the calculation amount needs to be increased by ten times, and the final following effect is basically unchanged, but if the precision requirement is reduced, it will directly lead to a significant decrease in the following effect.

[0092] Preferably, the (E aim , N aim ) is obtained by the following formula (five):

[0093] (g e (i)-E aim ) 2 +(g n (i)-N aim ) 2 =l d 2 (five).

[0094] The formula (five) is a process of continuously drawing a circle, and the comprehensive deviation is obtained by simultaneously solving the formula (five) and the above formula (two), (three), (four).

[0095] In a preferred embodiment, the weights and w1, w2 are obtained in real time through the following steps:

[0096] Step a, at the rear wheel position (g) of the ground motorized platform e (i), g n (i) is the center of the circle, with the minimum aiming distance l dmin Detect points on a given planned trajectory within a circular region of radius .

[0097] Step b: If there are no points on the planned trajectory within the circular area, then w1 = w2 = 0.5;

[0098] If there are points on the planned trajectory within the circular area, find the point (E1, N1) with the smallest lateral deviation e1 from the center position among the points on the planned trajectory within the circular area, and find the point (E2, N2) with the smallest longitudinal deviation e2 from the center position among the points on the planned trajectory within the circular area. Calculate the minimum lateral deviation e. 1min and the minimum longitudinal deviation e 2min ;

[0099] When e 1min ≥e 2min At that time, w1 = 0.9, w2 = 0.1

[0100] When e 1min <e 2min At that time, w1 = 0.8, w2 = 0.2.

[0101] The above method can improve the reproduction accuracy of path following. The purpose of this method is to make the comprehensive deviation e approach zero. Although it is impossible to make e always approach zero under various practical constraints, this method fully considers the coupling factors of lateral deviation and longitudinal deviation, and can minimize e within the achievable range.

[0102] In a preferred embodiment, sensors are installed on the unmanned ground mobility platform to detect and obtain the platform's speed, rear wheel position coordinates, and heading angle in real time. Specifically, the unmanned ground mobility platform is equipped with a GPS receiver to obtain the speed and rear wheel position coordinates in real time, and an inertial navigation component to obtain the heading angle in real time. Preferably, the unmanned ground mobility platform may also be equipped with a lidar or landmark recognition camera, which, in conjunction with its work site, identifies reference points to obtain its own speed, position, and heading angle in real time.

[0103] Step 2: Obtain the front wheel steering angle δ(t) based on the current aiming distance value.

[0104] Preferably, in step 2, the front wheel steering angle δ(t) is obtained by the following formula (six):

[0105]

[0106] wherein δ(t) represents the front wheel steering angle, L represents the wheelbase of the ground mobile platform, i.e. the distance between the front wheel and the rear wheel, (g e , g n ) represents the rear wheel position coordinate, (g x , g y ) represents the preview point coordinate, θ(t) represents the heading angle, and l d represents the preview distance.

[0107] A current preview distance is obtained in each sampling period and substituted into formula (six) to obtain the front wheel steering angle of the sampling period, i.e. to obtain the moving direction of the unmanned ground mobile platform in the sampling period.

[0108] In step 3, the value of δ(t) is output to the front wheel steering actuator, and the front wheel steering is controlled accordingly.

[0109] In step 3, it is determined whether the front wheel steering angle δ(t) obtained in step 2 is greater than the steering limit value,

[0110] When the front wheel steering angle δ(t) is greater than the steering limit value, the steering limit value is input to the front wheel steering actuator,

[0111] When the front wheel steering angle δ(t) is less than or equal to the steering limit value, the front wheel steering angle δ(t) is input to the front wheel steering actuator.

[0112] The steering limit value is related to the hardware structure of the unmanned ground mobile platform, and is selected and set according to the structural parameters of the unmanned ground mobile platform, for example, it is set to δ max = 25°.

[0113] In a preferred embodiment, steps 1, 2 and 3 are performed once in each sampling period, and the sampling period is the time for the sensor to obtain information. In this application, the sampling period is preferably set to 0.05s.

[0114] Embodiment 1

[0115] A raw planning path is given, i.e. a raw given path, and the unmanned ground mobile platform is controlled to follow the path, as shown in Figure 1 、 Figure 3 and Figure 5 .

[0116] Embodiment 1-1

[0117] The unmanned ground mobile platform is controlled to follow the path by the following method:

[0118] Step 1, call the fixed current preview distance value, l d = 3.7m;

[0119] Step 2, obtain the front wheel turning angle δ(t) according to the current preview distance value,

[0120] Step 3, output the δ(t) value to the front wheel turning execution mechanism, and control the front wheel turning according to the value.

[0121] The obtained following path trajectory is shown in Figure 1 , the lateral deviation change image in Figure 1 is shown in Figure 2 , and the horizontal coordinate in Figure 2 represents the time i, i.e. the calculation times.

[0122] Embodiment 1-2

[0123] The unmanned ground mobile platform is controlled to follow the path by the following method:

[0124] Step 1, obtain the current preview distance in real time,

[0125] Step 2, obtain the front wheel turning angle δ(t) according to the current preview distance value,

[0126] Step 3, output the δ(t) value to the front wheel turning execution mechanism, and control the front wheel turning according to the value.

[0127] In step 1, the value of the preview distance l d is traversed and substituted into formula (I), and the preview distance l d that makes the absolute value of the comprehensive deviation e(i+1) at the i+1 time minimum is selected as the current preview distance;

[0128] e(i+1) = K1(i+1) + f(v) + f(v, l d ) (I)

[0129] K1(i+1) = w1(g e (i) - N(i+1)) + w2(g n (i) - E(i+1)) (II)

[0130] f(v) = w1vt0 cos θ(i) + w2vt0 sin θ(i) (III)

[0131]

[0132] (g e (i) - E aim )2 +(g n (i)-N aim ) 2 =l d 2 (five)

[0133] Among them, the pre-aiming distance l d The value range is 0.5m ≤ l d ≤5m.

[0134] The weights and w1, w2 are obtained in real time through the following steps:

[0135] Step a, at the rear wheel position (g) of the ground motorized platform e (i), g n (i) is the center of the circle, with the minimum aiming distance l dmin Detect points on a given planned trajectory within a circular region of radius .

[0136] Step b: If there are no points on the planned trajectory within the circular area, then w1 = w2 = 0.5;

[0137] If there are points on the planned trajectory within the circular area, find the point (E1, N1) with the smallest lateral deviation e1 from the center position among the points on the planned trajectory within the circular area, and find the point (E2, N2) with the smallest longitudinal deviation e2 from the center position among the points on the planned trajectory within the circular area. Calculate the minimum lateral deviation e. 1min and the minimum longitudinal deviation e 2min ;

[0138] When e 1min ≥e 2min At that time, w1 = 0.9, w2 = 0.1

[0139] When e 1min <e 2min At that time, w1 = 0.8, w2 = 0.2.

[0140] The obtained following path trajectory is as follows Figure 3 As shown, Figure 3 The image of lateral deviation change in the image is as follows Figure 4 As shown, Figure 4 The horizontal axis represents time i, which is the number of calculations.

[0141] Examples 1-3

[0142] The unmanned ground mobile platform is controlled using a method essentially the same as in Examples 1-2, except that sub-step 1 only considers lateral deviation and does not assign weights.

[0143] Accordingly, the formula (I'), formula (II'), formula (III') and formula (IV') are adjusted as follows:

[0144] e'(i+1)=K1'(i+1)+F(v)+F(v,l d ) (I')

[0145] K1'(i+1)=g e (i)-N(i+1) (II')

[0146] F(v)=vt0 cos θ(i) (III')

[0147]

[0148] The obtained following path trajectory is shown in Figure 5 , and the lateral deviation change image in Figure 5 is shown in Figure 6 , and the horizontal coordinate in Figure 6 represents time i, i.e. the calculation number.

[0149] Example 2,

[0150] Given an original planning path, i.e. the original given path, the unmanned ground mobile platform is controlled to follow the path, as shown in Figure 7 , Figure 9 and Figure 11 .

[0151] Example 2-1

[0152] The unmanned ground mobile platform is controlled to follow the path by the following method:

[0153] Step 1, the fixed current preview distance value, l d = 1.3 m is called;

[0154] Step 2, the front wheel steering angle δ(t) is obtained according to the current preview distance value,

[0155] Step 3, the δ(t) value is output to the front wheel steering execution mechanism, and the front wheel steering is controlled accordingly.

[0156] The obtained following path trajectory is shown in Figure 7 , and the lateral deviation change image in Figure 7 is shown in Figure 8 , and the horizontal coordinate in Figure 8 represents time i, i.e. the calculation number.

[0157] Example 2-2

[0158] The unmanned ground mobile platform is controlled to follow the path by the following method:

[0159] Step 1, real-time obtain current preview distance,

[0160] Step 2, according to current preview distance value obtain front wheel rotation angle δ(t),

[0161] Step 3, output δ(t) value to front wheel steering actuator, according to which control front wheel steering.

[0162] Wherein, in step 1, traverse preview distance l d Value and substitute into formula (I), select the preview distance l d That makes the absolute value of the i+1 time comprehensive deviation e(i+1) minimum, as the current preview distance;

[0163] e(i+1) = K1(i+1) + f(v) + f(v, l d ) (I)

[0164] K1(i+1) = w1(g e (i) - N(i+1)) + w2(g n (i) - E(i+1)) (II)

[0165] f(v) = w1vt0 cos θ(i) + w2vt0 sin θ(i) (III)

[0166]

[0167] (g e (i) - E aim ) 2 +(g n (i) - N aim ) 2 =l d 2 (V)

[0168] Wherein, the preview distance l d The value range is 0.5m≤l d ≤5m.

[0169] The weights w1, w2 are obtained in real time by the following steps:

[0170] Step a, detect the point on the given planning trajectory in the circular region with the rear wheel position (g e (i), g n (i)) of the ground mobile platform as the center and the minimum preview distance l dmin As the radius;

[0171] Step b, if there is no point on the planning trajectory in the circular region, then w1 = w2 = 0.5;

[0172] If there are points on the planned trajectory within the circular area, find the point (E1, N1) with the smallest horizontal deviation e1 from the center of the circle among the points on the planned trajectory located within the circular area, and find the point (E2, N2) with the smallest vertical deviation e2 from the center of the circle among the points on the planned trajectory located within the circular area, and calculate the minimum horizontal deviation e 1min and the minimum vertical deviation e 2min ;

[0173] When e 1min ≥e 2min , w1 = 0.9, w2 = 0.1

[0174] When e 1min <e 2min , w1 = 0.8, w2 = 0.2.

[0175] The obtained following path trajectory is as shown in Figure 9 , Figure 9 The horizontal deviation change image in Figure 10 is as shown in Figure 10 where the abscissa in

[0176] Example 2 - 3

[0177] The unmanned ground mobile platform is controlled by a method substantially the same as that in Example 2 - 2, except that only the horizontal deviation is considered in sub - step 1 and no weights are set, <00​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​Figure 4 , Figure 6 The comparison Figure 8 and Figure 10 , Figure 12 The comparison shows that using a real-time adjusted and changing pre-aiming distance for control significantly improves control accuracy compared to the traditional method of using a fixed pre-aiming distance.

[0185] pass Figure 4 and Figure 6 The comparison Figure 10 and Figure 12 The comparison shows that by comprehensively considering both lateral and longitudinal deviations during the control process, the control accuracy can be further improved compared to the method that only considers lateral deviation.

[0186] The present invention has been described above with reference to preferred embodiments; however, these embodiments are merely exemplary and illustrative. Various substitutions and modifications can be made to the present invention based on these embodiments, all of which fall within the scope of protection of the present invention.

Claims

1. A path following control method applied to an unmanned ground mobile platform, characterized in that, the method comprises the following steps: Step 1, obtaining the current preview distance in real time, Step 2, obtaining the front wheel steering angle δ(t) according to the current preview distance value, Step 3, outputting the δ(t) value to the front wheel steering actuator, thereby controlling the front wheel steering; Steps 1, 2 and 3 are executed once in each sampling period; Step 1 comprises the following sub-steps: Sub-step 1, traverse the preview distance l d and substitute into formula (I), each l d value corresponds to a comprehensive deviation e(i+1) at the i+1 time. Sub-step 2, select the preview distance l that makes the absolute value of the integrated deviation e(i+1) at the i+1 moment minimum d is the current preview distance The formula (I) is: e(i+1) = K1(i+1) + f(v) + f(v, I d ) (one) Wherein, e(i+1) represents the integrated deviation at the i+1 time; K1(i+1) is obtained by the following equation (two), f(v) is obtained by the following equation (three), f(v, l d ) is obtained by the following equation (four); K1(i+1) = w1(g e (i)-N(i+1))+ w2(g n (i)-E(i+1)) (two) f(v) = w1vt0cosθ(i) + w2vt0sinθ(i) (III) Wherein, w1 represents the weight of the lateral deviation, and w2 represents the weight of the longitudinal deviation; (g e (i), g n (i)) represents the position coordinates of the rear wheel at the i-th time instant; (E(i+1), N(i+1)) represents the position coordinate (g e (i+1), g n (i+1) nearest point; v represents the speed of the unmanned ground mobile platform; t0 represents the sampling period; θ(i) represents the heading angle at the i time; (E aim , N aim ) represents the i-th moment preview point coordinates; l d represents a preview distance.

2. The path following control method applied to the unmanned ground mobile platform according to claim 1, characterized in that, In sub-step 1, the pre-aim distance l d is in the range of 0.5m≤l d ≤5m; the value precision of l d is 0.1m during the value traversal process.

3. The path following control method applied to the unmanned ground mobile platform according to claim 1, characterized in that, The weights w1 and w2 are obtained in real time by the following steps: Step a, detecting a point on the given planned trajectory within a circular region centered at the rear wheel position (g e (i), g n (i)) of the ground mobile platform with a minimum preview distance l dmin as a radius; Step b, if there is no point on the planned trajectory in the circular region, then w1 = w2 = 0.5; If there are points on the planned trajectory within the circular area, find the point (E1, N1) with the smallest lateral deviation e1 from the center position among the points on the planned trajectory within the circular area, and find the point (E2, N2) with the smallest longitudinal deviation e2 from the center position among the points on the planned trajectory within the circular area. Calculate the minimum lateral deviation e. 1min and the minimum longitudinal deviation e 2min ; When e 1min ≥ e 2min w1 = 0.9, w2 = 0.1 When e 1min <e 2min w1 = 0.8, w2 = 0.

2.

4. The path following control method applied to the unmanned ground mobile platform according to claim 1, characterized in that, A sensor is arranged on the unmanned ground mobile platform to detect the speed, rear wheel position coordinates and heading angle of the unmanned ground mobile platform in real time.

5. The path following control method applied to the unmanned ground mobile platform according to claim 1, characterized in that, said (E aim , N aim ) is obtained by the following formula (five): (g e (i)-E aim ) 2 +(g n (i)-N aim ) 2 = l d 2 (five).

6. The path following control method applied to the unmanned ground mobile platform according to claim 1, characterized in that, In step 2, the front wheel steering angle δ(t) is obtained by the following formula (VI): wherein δ(t) represents the front wheel steering angle, L represents the wheelbase of the ground mobile platform, (g e , g n ) represents the rear wheel position coordinates, (g x , g y ) represents the preview point coordinates, θ(t) represents the heading angle, and l d represents the preview distance.

7. The path following control method applied to the unmanned ground mobile platform according to claim 1, characterized in that, In step 3, it is judged whether the front wheel steering angle δ(t) obtained in step 2 is greater than the steering extreme value, When the current front wheel steering angle δ(t) is greater than the steering extreme value, the steering extreme value is input to the front wheel steering actuator, When the current front wheel steering angle δ(t) is less than or equal to the steering extreme value, the current front wheel steering angle δ(t) is input to the front wheel steering actuator.

Images