CN116279792A - Vehicle-mounted intelligent steering control system - Google Patents
Vehicle-mounted intelligent steering control system Download PDFInfo
- Publication number
- CN116279792A CN116279792A CN202310574408.9A CN202310574408A CN116279792A CN 116279792 A CN116279792 A CN 116279792A CN 202310574408 A CN202310574408 A CN 202310574408A CN 116279792 A CN116279792 A CN 116279792A
- Authority
- CN
- China
- Prior art keywords
- distance
- real
- time
- reserved
- nonlinear model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000012937 correction Methods 0.000 claims description 30
- 238000000034 method Methods 0.000 claims description 22
- 230000007613 environmental effect Effects 0.000 claims description 19
- 238000012216 screening Methods 0.000 claims description 5
- 238000007405 data analysis Methods 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 abstract description 12
- 238000000605 extraction Methods 0.000 abstract description 3
- 230000000875 corresponding effect Effects 0.000 description 58
- 230000000694 effects Effects 0.000 description 6
- 230000001276 controlling effect Effects 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 230000001186 cumulative effect Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000002708 enhancing effect Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000012954 risk control Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B62—LAND VEHICLES FOR TRAVELLING OTHERWISE THAN ON RAILS
- B62D—MOTOR VEHICLES; TRAILERS
- B62D6/00—Arrangements for automatically controlling steering depending on driving conditions sensed and responded to, e.g. control circuits
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Steering Control In Accordance With Driving Conditions (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention relates to the technical field of control systems, in particular to a vehicle-mounted intelligent steering control system, which obtains an initial nonlinear model of a reference distance and expected steering resistance of an operating rod according to the rotating direction of the operating rod of an engineering vehicle, the reference distance and a preset risk coefficient; obtaining the availability degree of each distance threshold according to the data distribution characteristics of the distance histogram corresponding to the actual distance, and selecting a reserved distance; and carrying out feature extraction analysis on the reserved distance to obtain a real-time collision risk coefficient, updating the initial nonlinear model to obtain a final nonlinear model, and finally realizing steering control on engineering vehicle operation. The invention can reduce the error of real-time steering resistance control, ensure the safety of engineering vehicles and improve the working efficiency.
Description
Technical Field
The invention relates to the technical field of control systems, in particular to a vehicle-mounted intelligent steering control system.
Background
During the operation of the engineering vehicle, the operating lever of the engineering vehicle needs to be continuously rotated, and therefore, collision may occur, resulting in equipment damage. The existing engineering vehicle vision system can be used for carrying out three-dimensional modeling on the surrounding environment of the engineering vehicle, further controlling the operation of the engineering vehicle according to the actual distance between the real-time engineering vehicle and surrounding objects, and preventing the engineering vehicle staff from being impacted and endangering the safety of the engineering vehicle staff in the operation process.
In the prior art, an alarm is given to engineering vehicle operators through the distance between the current engineering vehicle and surrounding environmental objects, different boundary distances and alarm time periods are preset directly according to the operating range of an operating lever of the engineering vehicle, the position of the passing obstacle is judged, the engineering vehicle is moved to avoid the obstacle, and the obstacle is avoided without difference, so that the possibility of collision of the engineering vehicle is avoided to a great extent, and the working efficiency of the engineering vehicle is affected. Or, the engineering vehicle presets the preset value of the gesture of the bucket relative to the bucket rod for the control rod, the preset gesture is set only according to experience, the emergency possibly encountered in the operation process is not considered, the scientificity is lacking, and the control error can be increased.
Disclosure of Invention
In order to solve the technical problems that in the prior art, the analysis of steering control of a joystick is not combined with emergency, so that operation control errors are increased and working efficiency is reduced, the invention aims to provide a vehicle-mounted intelligent steering control system, and the adopted technical scheme is as follows:
the invention provides a vehicle-mounted intelligent steering control system, which comprises:
the preset model module is used for obtaining an initial nonlinear model of the reference distance and the expected steering resistance of the operating rod according to the rotating direction of the operating rod of the engineering vehicle, the reference distance and a preset risk coefficient; the reference distance is the distance between the engineering vehicle and the environmental object;
the real-time data module is used for obtaining at least two real-time environment images of the engineering vehicle and obtaining the actual distance between the engineering vehicle and each environment object in each real-time environment image respectively; constructing corresponding distance histograms based on all actual distances, and taking all abscissa values of each distance histogram as distance threshold values; obtaining the availability degree of each distance threshold according to the data distribution characteristics of the distance histogram; screening out the reserved distance in the corresponding real-time environment image according to the availability degree of all the distance thresholds;
the data analysis module is used for classifying all the reserved distances according to each reserved distance and the position information of the corresponding reserved distance in the corresponding real-time environment image to obtain at least two distance categories; respectively obtaining real-time distances and real-time collision risk coefficients according to the distance characteristics of all the distance categories;
the real-time control module is used for updating the initial nonlinear model according to the real-time collision risk coefficient to obtain a final nonlinear model; obtaining expected steering resistance according to the real-time distance and the final nonlinear model; acquiring real-time steering resistance of the engineering vehicle; and controlling the working condition of the engineering vehicle according to the relation between the real-time steering resistance and the expected steering resistance.
Further, the method for acquiring the nonlinear model of the reference distance and the expected steering resistance of the joystick comprises the following steps:
dividing the expected rotation direction of the control lever into a power-assisted forward direction and a power-assisted reverse direction according to the rotation direction of the control lever;
the power-assisted reversing of the rotating direction of the operating lever takes the product of the reference distance and the initial risk coefficient as a reversing feature, and a reversing nonlinear model of the reference distance and the expected steering resistance of the operating lever is constructed according to the reversing feature by the first adjusting parameter and the standard normal model;
the method comprises the steps of obtaining a correction ratio of a preset correction coefficient to an initial risk coefficient for the forward direction of the assistance force of the rotating direction of the operating lever, and taking the sum of the correction ratio and the preset correction coefficient as a correction parameter; taking the difference between the reference distance and the correction parameter as a forward characteristic; constructing a forward nonlinear model of the reference distance and the expected steering resistance of the control rod according to the second adjustment parameter and the standard normal model;
the initial nonlinear model of the reference distance and the expected steering resistance of the joystick includes a reverse nonlinear model and a forward nonlinear model of the reference distance and the expected steering resistance of the joystick.
Further, the method for acquiring the availability degree of the distance threshold value comprises the following steps:
for any one distance threshold, taking other distance thresholds which are not more than the distance threshold as reference distance thresholds of the corresponding distance thresholds, and obtaining the quantity ratio of each reference distance threshold of each distance threshold in the distance histogram; traversing all the reference distance thresholds of each distance threshold from large to small, and taking the ratio of the number ratio of each reference distance threshold to the corresponding traversing sequence as the sub-availability of the corresponding reference distance threshold; taking the accumulated average value of all the sub-availability values of the reference distance threshold values of each distance threshold value as an availability coefficient; and taking the product of the preset correction parameter and the available coefficient as the available degree of the corresponding distance threshold value.
Further, the method for screening the reserved distance in the real-time environment image comprises the following steps:
for any one real-time environment image, taking a distance threshold corresponding to the maximum value in the available degrees of all the distance thresholds as a reserved threshold;
and for all the actual distances corresponding to any one real-time environment image, if the actual distances are smaller than or equal to the reserved threshold value, the actual distances are reserved distances.
Further, the method for acquiring the real-time distance and the real-time collision risk coefficient comprises the following steps:
taking the distance average value of all the reserved distance values contained in each distance category as the distance characteristic of the corresponding distance category; taking the distance category corresponding to the minimum distance feature in all the distance categories as a real-time distance category, and taking the minimum reserved distance in the real-time distance category as a real-time distance;
and taking the ratio of the preset performance coefficient to the real-time distance as a real-time collision risk coefficient.
Further, the method for acquiring the real-time steering resistance comprises the following steps:
and obtaining a real-time torque value in an electric power steering system of the engineering vehicle, and obtaining corresponding real-time steering resistance according to the real-time torque value.
The invention has the following beneficial effects:
in the embodiment of the invention, the initial nonlinear model of the reference distance and the expected steering resistance of the operating lever is obtained first, the operating lever rotation direction of the engineering vehicle, the reference distance and the preset risk coefficient are correlated, and the initial nonlinear model is determined, so that the accuracy of subsequent analysis can be enhanced, and the error of the final nonlinear model is reduced. The available degree of each distance threshold is obtained according to the data distribution characteristics of the distance histogram corresponding to the actual distance, the reserved distance is selected according to the available degree of all the distance threshold, the unsuitable actual distance is screened out, the reserved distance is subjected to subsequent analysis, and the reserved distance can be accurately identified, so that errors are reduced. And classifying all the reserved distances according to the classification characteristics of each reserved distance, so as to obtain the real-time distance and the real-time collision risk coefficient, ensure the safety of the engineering vehicle in real-time operation and reduce the subsequent error. And updating the initial nonlinear model according to the real-time collision risk coefficient to obtain a final nonlinear model, obtaining expected steering resistance, enhancing the accuracy of the final nonlinear model, enhancing the accuracy of the expected steering resistance and reducing the working safety risk of the engineering vehicle. The working condition of the engineering vehicle is controlled according to the real-time steering resistance and the expected steering resistance, so that the safety of the engineering vehicle can be ensured and the working efficiency can be improved. According to the invention, an initial nonlinear model of the reference distance and the expected steering resistance is constructed, the actual distance between the engineering vehicle and surrounding objects is screened, the screened reserved distance is subjected to feature extraction analysis to obtain a real-time collision risk coefficient, the initial nonlinear model is updated to obtain a final nonlinear model, steering control of the engineering vehicle operation is finally realized, the error of steering resistance control is reduced, the safety of the engineering vehicle can be ensured, and the working efficiency is improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a block diagram of a vehicle-mounted intelligent steering control system according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description refers to the specific implementation, structure, characteristics and effects of a vehicle-mounted intelligent steering control system according to the invention with reference to the accompanying drawings and the preferred embodiment. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of the vehicle-mounted intelligent steering control system provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a block diagram of a vehicle-mounted intelligent steering control system according to an embodiment of the present invention is shown, where the system includes:
the preset model module S1 is used for obtaining an initial nonlinear model of the reference distance and the expected steering resistance of the operating rod according to the rotating direction of the operating rod of the engineering vehicle, the reference distance and a preset risk coefficient; the reference distance is the distance between the engineering vehicle and the environmental object.
In the working process of the engineering vehicle, if surrounding objects are located in an original working operation path of the engineering vehicle, a control lever of the engineering vehicle collides with the surrounding objects, the control of steering resistance of the control lever can influence the severity of the collision, and hidden danger of collision exists.
In order to reduce the operation risk of the engineering vehicle, reduce the risk of collision to the engineering vehicle to the greatest extent, judge the expected steering resistance of the operating lever, firstly, judge the relation between the distance between the engineering vehicle and the environmental object, the preset risk coefficient and the rotating direction of the operating lever, determine the initial nonlinear model of the reference distance and the expected steering resistance of the operating lever, wherein the reference distance is the distance between the engineering vehicle and the environmental object, and enhance the accuracy of the subsequent analysis, thereby reducing the error of the final nonlinear model. Obtaining an initial nonlinear model of the reference distance and the expected steering resistance of the operating rod according to the rotating direction of the operating rod of the engineering vehicle, the reference distance and a preset risk coefficient, wherein the initial nonlinear model specifically comprises the following steps:
the electric power steering system in the engineering vehicle can determine the rotation direction of the motor and the magnitude of the power-assisted current according to signals of the vehicle speed sensor and the torque sensor, thereby completing the real-time control of the power-assisted steering and realizing the different power-assisted effects of the motor when the vehicle speeds are different. The expected rotation direction is judged according to the difference between the rotation direction of the motor and the rotation direction of the operating lever of the engineering vehicle, and the accuracy of the result can be improved. Thus, the intended rotation direction of the joystick is divided into a power assist forward direction and a power assist reverse direction according to the rotation direction of the joystick, and in the embodiment of the present invention, specifically includes:
the smaller the reference distance is, the smaller the distance between the engineering vehicle and the environmental object is, the greater the possibility that the engineering vehicle collides with the corresponding environmental object is, and the higher the difficulty of operating the body operating lever by an operator is. For the safety of the engineering vehicle and operators, the motor rotation direction in the electric power steering system of the equipment should be opposite to the rotation direction of the engineering vehicle operating lever, so as to reversely assist the operating lever, and the expected rotation direction of the operating lever is the power assist reverse direction.
The larger the reference distance is, the larger the distance between the engineering vehicle and the environmental object is, the smaller the possibility that the engineering vehicle collides with the corresponding environmental object is, and the lower the difficulty of operating the body operating lever by operators is. Also, for the work efficiency and safety of the construction vehicle, the motor rotation direction in the electric power steering system of the apparatus should be consistent with the rotation direction of the operation lever of the construction vehicle, and forward power is applied to the operation lever, and at this time, the expected rotation direction of the operation lever is the power-assisted forward direction.
In order to enhance the accuracy of the subsequent control of the real-time steering resistance and reduce the operation error of the engineering vehicle, the preset steering resistance of the operating lever is firstly determined, the relation among the reference distance, the preset risk coefficient and the preset rotation direction is determined, and an initial nonlinear model of the reference distance and the expected steering resistance of the operating lever is constructed. Since the expected rotational direction of the joystick is divided into the assist forward direction and the assist reverse direction, the initial nonlinear model also needs to be distinguished according to the difference in the expected rotational direction. The normal distribution accords with the rules of most of the nature, so in the embodiment of the invention, the combination of two standard normal models is selected to be used as an initial nonlinear model. The initial nonlinear model of the reference distance and the expected steering resistance of the joystick includes a reverse nonlinear model and a forward nonlinear model of the reference distance and the expected steering resistance of the joystick, specifically including:
and reversing the power of the rotating direction of the operating lever, taking the product of the reference distance and the initial risk coefficient as a reversing feature, and constructing a reversing nonlinear model of the reference distance and the expected steering resistance of the operating lever according to the reversing feature by the first adjusting parameter and the standard normal model, wherein the reversing nonlinear model specifically comprises:
in the case of the inverse non-linear model,
representing a reverse non-linear model of the model,
representing an image of the real-time environment,
a first adjustment parameter is indicated and a second adjustment parameter is indicated,
representing the initial risk factor(s),
the reference distance is indicated as such,
the reverse feature is indicated and is shown,
representing the reverse feature
Is subject to a standard normal distribution of the distribution,
is a reverse feature
Is a function of the corresponding function of (a). In the embodiment of the present invention, the first adjustment parameter is 3.5, and the value of the first adjustment parameter may be specifically set according to the specific implementation manner. The second adjustment parameter is to adjust the value range of the model.
The method comprises the steps of obtaining a correction ratio of a preset correction coefficient to an initial risk coefficient for the forward direction of the assistance force of the rotating direction of the operating lever, and taking the sum of the correction ratio and the preset correction coefficient as a correction parameter; taking the difference between the reference distance and the correction parameter as a forward characteristic; and constructing a forward nonlinear model of the reference distance and the expected steering resistance of the control rod according to the second adjustment parameter and the standard normal model, wherein the forward nonlinear model specifically comprises the following components:
in the case of a forward non-linear model,
representing a reverse non-linear model of the model,
representing an image of the real-time environment,
a second adjustment parameter is indicated and is indicated,
the preset correction coefficient is indicated to be the same,
representing the initial risk factor(s),
the reference distance is indicated as such,
indicating that the corrected ratio value is to be displayed,
the correction coefficient is represented by a number of coefficients,
the positive-going characteristic is represented by a positive-going characteristic,
representing forward features
Is subject to a standard normal distribution of the distribution,
is a positive characteristic
Is a function of the corresponding function of (a). In the embodiment of the present invention, the second adjustment parameter is 3.5, the preset correction coefficient is 4.5, and the values of the second adjustment parameter and the preset correction coefficient can be specifically set according to the specific implementation manner. The preset correction coefficient has the function of moving the position with the normal distribution of 0 and simplifying the forward nonlinear model. The second adjustment parameter is to adjust the value range of the model.
Thus, through the design of steps A and B, the initial nonlinear model of the reference distance and the expected steering resistance of the joystick is
. In the embodiment of the invention, the real-time collision risk coefficient is also required to be calculated, and then the steering resistance is expectedAnd replacing the initial risk coefficient, and updating the initial nonlinear model to improve the model accuracy.
The real-time data module S2 is used for obtaining at least two real-time environment images of the engineering vehicle and obtaining the actual distance between the engineering vehicle and each environment object in each real-time environment image respectively; constructing corresponding distance histograms based on all actual distances, and taking all abscissa values of each distance histogram as distance threshold values; obtaining the availability degree of each distance threshold according to the data distribution characteristics of the distance histogram; and screening the reserved distance in the corresponding real-time environment image according to the availability degree of all the distance thresholds.
And installing a visual panoramic system around the body of the engineering vehicle, wherein the panoramic visual system comprises a depth camera, extracting surrounding environment images of the engineering vehicle through the depth camera, obtaining the distance between the real-time engineering vehicle and each environment object, and taking the obtained environment images as real-time environment images.
The front half part of the engineering vehicle body does not need to establish a panoramic visualization system, and the shape of the whole surface of the rear half part of the engineering vehicle body is rectangular, so that the depth camera is arranged on two sides of the rectangular corner angle of the rear half part of the engineering vehicle body. Therefore, in the embodiment of the invention, D depth cameras are selected to be installed, and for any real-time state, the D depth cameras can obtain a corresponding real-time environment image. The real-time environment images comprise the actual distance between the engineering vehicle and each environment object, and each real-time environment image is analyzed to obtain a corresponding real-time collision risk coefficient. The specific installation location practitioner may be set according to the specific number of cameras and loader body size. The corresponding distance units in the real-time environment image are millimeters, and the specific distance units can be specifically set according to specific acquisition equipment.
If data noise exists in the real-time environment image obtained by the depth camera, and the final distance value is determined only by the minimum distance value in the real-time environment image, a larger error exists in the final result. The existing denoising algorithm cannot guarantee the stability of each denoising effect. The position distribution of each environmental object contained in the real-time environmental image is discrete, and if clustering is performed to find a category containing the minimum value of the actual distance, the efficiency is reduced. As can be seen from the block S1, in the design of the initial nonlinear model, the reference distance is the distance between the engineering vehicle and the environmental object, and the reference distance is inversely related to the possibility of collision of the engineering vehicle. In practical application, the actual distance between the engineering vehicle and each environmental object in a real-time state is obtained according to the real-time environmental image, the actual distance can be regarded as a reference distance in the initial nonlinear model, and the smaller the actual distance is, the closer the distance between the engineering vehicle and the environmental object is, the greater the possibility that the engineering vehicle collides with the corresponding environmental object is, namely the higher the collision risk coefficient is. Therefore, the obtained actual distance is analyzed later, so that the accuracy of the result is enhanced, and the error is reduced.
Therefore, corresponding distance histograms are constructed based on all actual distances in the real-time environment image, and all abscissa values of each distance histogram are used as distance threshold values. In practical application, when the actual distance between the engineering vehicle and an environmental object exceeds a certain range, the possibility of collision between the engineering vehicle and the environmental object is extremely low, the actual distance exceeding the range is regarded as an invalid distance, and the invalid distance is screened out in analysis, so that the obtained collision risk coefficient is more accurate. In order to further distinguish between the effective collision distance and the ineffective collision distance, it is necessary to determine the availability of the distance threshold corresponding to the actual distance, and the greater the availability, the greater the likelihood that the actual distance corresponding to the distance threshold is the ineffective collision distance. Thus, the available degree of each distance threshold is obtained according to the data distribution characteristics of the distance histogram, which specifically includes:
for any one distance threshold, taking other distance thresholds which are not more than the distance threshold as reference distance thresholds of the corresponding distance thresholds, and obtaining the quantity ratio of each reference distance threshold of each distance threshold in the distance histogram; traversing all the reference distance thresholds of each distance threshold from large to small, and taking the ratio of the number ratio of each reference distance threshold to the corresponding traversing sequence as the sub-availability of the corresponding reference distance threshold; taking the accumulated average value of all the sub-availability values of the reference distance threshold values of each distance threshold value as an availability coefficient; and taking the product of the preset correction parameter and the available coefficient as the available degree of the corresponding distance threshold value. The availability formula specifically comprises:
in the method, in the process of the invention,
represent the first
The degree of availability of the individual distance thresholds,
representing the number of distance thresholds that are to be used,
representing the first corresponding distance threshold
The number of reference distance thresholds is chosen such that,
represent the first
The number of reference distance thresholds is a ratio,
representing a preset correction parameter.
In the availability formula, the first
The number of reference distance threshold values is counted and the corresponding traversal order is calculated
Is compared with the ratio of (1)
The larger the distance threshold value selected in the real-time environment image is the minimum value, the larger the corresponding threshold value availability degree is, namely the more accurate the effect of distinguishing the effective collision distance from the ineffective collision distance is;
is the first
The number of reference distance thresholds is a ratio,
the larger the number of the corresponding reference distance threshold value in the distance histogram is, the more accurate the clustering effect is when the reserved distances contained in the corresponding reference distance threshold value are clustered in subsequent analysis.
Represent the first
The greater the sub-availability of the reference distance threshold value, the greater the availability of the corresponding reference distance threshold value, i.e., the greater the likelihood that the actual distance corresponding to the reference distance threshold value is an invalid collision distance.
Represent the first
The individual distance thresholds correspond to the cumulative average of the sub-availability of all reference distance thresholds, the greater the cumulative average indicating the first
The greater the availability of the distance threshold, i.e. the first
The less likely the actual distance corresponding to the individual distance thresholds is an invalid collision distance. The preset correction parameter can adjust the numerical range of the formula result, and in the embodiment of the invention, the preset correction parameter is 1, and the numerical value of the specific preset correction parameter can be specifically set according to the specific implementation mode.
The availability degree of the distance threshold value indicates the possibility that the actual distance corresponding to the distance threshold value is an invalid collision distance, and if the availability degree is larger, the possibility that the actual distance corresponding to the distance threshold value is an invalid collision distance is larger, namely the association degree between the actual distance corresponding to the distance threshold value and the collision is smaller; the smaller the availability, the less likely the actual distance corresponding to the distance threshold is an invalid collision distance, i.e., the greater the degree of association of the actual distance corresponding to the distance threshold and the collision.
The possibility that the actual distance corresponding to the distance threshold is an invalid collision distance can be judged according to the availability degree of each distance threshold, and in order to further screen the effective collision distance from the actual distances, the actual distance with a larger association degree with the collision is obtained, and the reserved distance in the corresponding real-time environment image is screened according to the availability degree of all the distance thresholds, which specifically comprises the following steps:
and regarding any one real-time environment image, taking the distance threshold corresponding to the maximum value in the available degrees of all the distance thresholds as a reserved threshold. And for all the actual distances corresponding to any one real-time environment image, if the actual distances are smaller than or equal to the reserved threshold value, the actual distances are reserved distances.
And (3) obtaining the reserved distances in all the real-time environment images, namely the effective collision distances, through analysis of the module S2.
The data analysis module S3 is used for classifying all the reserved distances according to each reserved distance and the position information of the corresponding reserved distance in the corresponding real-time environment image to obtain at least two distance categories; and respectively obtaining the real-time distance and the real-time collision risk coefficient according to the distance characteristics of all the distance categories.
In order to distinguish the reserved distances of each real-time environment image and enhance the working efficiency of the engineering vehicle, the position information of the reserved distances in the corresponding real-time environment images needs to be obtained, and all the reserved distances are classified according to the position information of each reserved distance and the corresponding reserved distance in the corresponding real-time environment images to obtain at least two distance categories. In the embodiment of the invention, the sample distance is obtained by calculating the difference distance between the position information among the reserved distances and the reserved distance value, each reserved distance is mapped to the sample space according to the sample distance, the reserved distance is classified by using a DBSCAN clustering algorithm, the clustering radius is set to 3, the density threshold is set to 3, the values of the specific clustering radius and the density threshold can be specifically set according to specific embodiments, and in other implementation scenes, an implementer can select other classification methods such as a k-means algorithm according to specific scenes. It should be noted that, the DBSCAN clustering algorithm is a technical means well known to those skilled in the art, and will not be described herein.
And analyzing the reserved distance contained in each distance category to obtain a corresponding distance characteristic, and analyzing the distance characteristic corresponding to each distance category to obtain the minimum effective collision distance between the engineering vehicle and the surrounding environment in a real-time state, namely, the real-time distance, so that the result error is reduced. Therefore, the real-time distance and the real-time collision risk coefficient are respectively obtained according to the distance characteristics of all the distance categories, and specifically comprise:
taking the distance average value of all the reserved distance values contained in each distance category as the distance characteristic of the corresponding distance category, taking the distance category corresponding to the minimum distance characteristic in all the distance categories as the real-time distance category, taking the minimum reserved distance in the real-time distance category as the real-time distance, and taking the ratio of the preset performance coefficient to the real-time distance as the real-time collision risk coefficient. In the embodiment of the present invention, the preset coefficient of performance is 70, and the numerical value of the specific preset coefficient of performance may be specifically set according to the specific implementation manner. The real-time distance and the real-time collision risk coefficient are in inverse proportion, and the smaller the real-time distance is, the smaller the minimum effective collision distance between the engineering vehicle and the surrounding environment in the real-time state is, and the greater the possibility of collision is, namely the greater the real-time collision risk coefficient is.
And (3) obtaining the real-time distance and the real-time collision risk coefficient of the engineering vehicle in a real-time state through analysis of the module S3.
The real-time control module S4 is used for updating the initial nonlinear model according to the real-time collision risk coefficient to obtain a final nonlinear model; obtaining expected steering resistance according to the real-time distance and the final nonlinear model; acquiring real-time steering resistance of the engineering vehicle; and controlling the working condition of the engineering vehicle according to the relation between the real-time steering resistance and the expected steering resistance.
And (3) obtaining an initial nonlinear model through analysis of the modules S1, S2 and S3, and determining the real-time distance and the real-time collision risk coefficient of the engineering vehicle in a real-time state.
Firstly, substituting the obtained real-time collision risk coefficient into an initial nonlinear model, and updating the initial nonlinear model to obtain a final nonlinear model. The real-time collision risk coefficient is changed according to the real-time distance, and the final nonlinear model is also dynamically changed.
Secondly, obtaining the real-time steering resistance of the engineering vehicle, which specifically comprises the following steps: and obtaining a real-time torque value in an electric power steering system of the engineering vehicle, and obtaining corresponding real-time steering resistance according to the real-time torque value. The operating lever of the engineering vehicle is connected with a torque sensor in the electric power steering system, and the real-time torque value of the torque sensor represents the real-time steering resistance condition of the operating lever of the engineering vehicle in a real-time state, and the method for obtaining the resistance according to the torque value of the torque sensor is a technical means well known to those skilled in the art and is not repeated herein.
And then, according to the real-time distance and the final nonlinear model, the expected steering resistance of the engineering vehicle operating lever in a real-time state is obtained. If the expected steering resistance is negative, the expected rotation direction of the operating lever is the power-assisted reverse direction, namely the rotation direction of a motor in an electric power-assisted steering system of the equipment is opposite to the rotation direction of the operating lever of the engineering vehicle; if the expected steering resistance is a positive value, the expected rotational direction of the joystick is the positive power assist direction, i.e. the motor rotational direction in the electric power steering system of the device should be kept consistent with the rotational direction of the engineering vehicle joystick.
Finally, the working condition of the engineering vehicle is controlled according to the relation between the real-time steering resistance and the expected steering resistance, and the method specifically comprises the following steps: the real-time steering resistance reaches the expected steering resistance by adjusting the current of the motor, so that the engineering vehicle operation control is completed, and the safety risk of operators is reduced. Wherein, the adjustment mode of motor current size: and obtaining the required rotating motor current according to the rotating motor current and the expected steering resistance relation table, and adjusting the real-time motor current to an expected value to finish the operation risk control of the engineering vehicle.
In summary, the embodiment of the invention firstly builds the initial nonlinear model of the reference distance and the expected steering resistance, screens the actual distance between the engineering vehicle and surrounding objects, performs feature extraction analysis on the screened reserved distance to obtain the real-time collision risk coefficient, further updates the initial nonlinear model to obtain the final nonlinear model, finally realizes the steering control of the engineering vehicle operation, reduces the error of the steering resistance control, and can ensure the safety of the engineering vehicle and improve the working efficiency.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
Claims (6)
1. An on-vehicle intelligent steering control system, characterized in that the system includes:
the preset model module is used for obtaining an initial nonlinear model of the reference distance and the expected steering resistance of the operating rod according to the rotating direction of the operating rod of the engineering vehicle, the reference distance and a preset risk coefficient; the reference distance is the distance between the engineering vehicle and the environmental object;
the real-time data module is used for obtaining at least two real-time environment images of the engineering vehicle and obtaining the actual distance between the engineering vehicle and each environment object in each real-time environment image respectively; constructing corresponding distance histograms based on all actual distances, and taking all abscissa values of each distance histogram as distance threshold values; obtaining the availability degree of each distance threshold according to the data distribution characteristics of the distance histogram; screening out the reserved distance in the corresponding real-time environment image according to the availability degree of all the distance thresholds;
the data analysis module is used for classifying all the reserved distances according to each reserved distance and the position information of the corresponding reserved distance in the corresponding real-time environment image to obtain at least two distance categories; respectively obtaining real-time distances and real-time collision risk coefficients according to the distance characteristics of all the distance categories;
the real-time control module is used for updating the initial nonlinear model according to the real-time collision risk coefficient to obtain a final nonlinear model; obtaining expected steering resistance according to the real-time distance and the final nonlinear model; acquiring real-time steering resistance of the engineering vehicle; and controlling the working condition of the engineering vehicle according to the relation between the real-time steering resistance and the expected steering resistance.
2. The vehicle-mounted intelligent steering control system according to claim 1, wherein the method for acquiring the nonlinear model of the reference distance and the expected steering resistance of the joystick comprises:
dividing the expected rotation direction of the control lever into a power-assisted forward direction and a power-assisted reverse direction according to the rotation direction of the control lever;
the power-assisted reversing of the rotating direction of the operating lever takes the product of the reference distance and the initial risk coefficient as a reversing feature, and a reversing nonlinear model of the reference distance and the expected steering resistance of the operating lever is constructed according to the reversing feature by the first adjusting parameter and the standard normal model;
the method comprises the steps of obtaining a correction ratio of a preset correction coefficient to an initial risk coefficient for the forward direction of the assistance force of the rotating direction of the operating lever, and taking the sum of the correction ratio and the preset correction coefficient as a correction parameter; taking the difference between the reference distance and the correction parameter as a forward characteristic; constructing a forward nonlinear model of the reference distance and the expected steering resistance of the control rod according to the second adjustment parameter and the standard normal model;
the initial nonlinear model of the reference distance and the expected steering resistance of the joystick includes a reverse nonlinear model and a forward nonlinear model of the reference distance and the expected steering resistance of the joystick.
3. The vehicle-mounted intelligent steering control system according to claim 1, wherein the method for acquiring the availability of the distance threshold comprises:
for any one distance threshold, taking other distance thresholds which are not more than the distance threshold as reference distance thresholds of the corresponding distance thresholds, and obtaining the quantity ratio of each reference distance threshold of each distance threshold in the distance histogram; traversing all the reference distance thresholds of each distance threshold from large to small, and taking the ratio of the number ratio of each reference distance threshold to the corresponding traversing sequence as the sub-availability of the corresponding reference distance threshold; taking the accumulated average value of all the sub-availability values of the reference distance threshold values of each distance threshold value as an availability coefficient; and taking the product of the preset correction parameter and the available coefficient as the available degree of the corresponding distance threshold value.
4. The vehicle-mounted intelligent steering control system according to claim 3, wherein the method for screening the reserved distance in the real-time environment image comprises the following steps:
for any one real-time environment image, taking a distance threshold corresponding to the maximum value in the available degrees of all the distance thresholds as a reserved threshold;
and for all the actual distances corresponding to any one real-time environment image, if the actual distances are smaller than or equal to the reserved threshold value, the actual distances are reserved distances.
5. The vehicle-mounted intelligent steering control system according to claim 1, wherein the method for acquiring the real-time distance and the real-time collision risk coefficient comprises:
taking the distance average value of all the reserved distance values contained in each distance category as the distance characteristic of the corresponding distance category; taking the distance category corresponding to the minimum distance feature in all the distance categories as a real-time distance category, and taking the minimum reserved distance in the real-time distance category as a real-time distance;
and taking the ratio of the preset performance coefficient to the real-time distance as a real-time collision risk coefficient.
6. The vehicle-mounted intelligent steering control system according to claim 1, wherein the method for acquiring the real-time steering resistance comprises:
and obtaining a real-time torque value in an electric power steering system of the engineering vehicle, and obtaining corresponding real-time steering resistance according to the real-time torque value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310574408.9A CN116279792B (en) | 2023-05-22 | 2023-05-22 | Vehicle-mounted intelligent steering control system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310574408.9A CN116279792B (en) | 2023-05-22 | 2023-05-22 | Vehicle-mounted intelligent steering control system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116279792A true CN116279792A (en) | 2023-06-23 |
| CN116279792B CN116279792B (en) | 2023-07-18 |
Family
ID=86785366
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310574408.9A Active CN116279792B (en) | 2023-05-22 | 2023-05-22 | Vehicle-mounted intelligent steering control system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116279792B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011128844A (en) * | 2009-12-17 | 2011-06-30 | Fuji Heavy Ind Ltd | Road shape recognition device |
| CN102601862A (en) * | 2012-04-01 | 2012-07-25 | 中联重科股份有限公司 | Engineering vehicle, alerting device and alerting method for safe steering of engineering vehicle |
| KR20130005110A (en) * | 2011-07-05 | 2013-01-15 | 현대자동차주식회사 | Method for controlling vehicle interval |
| JP2019128153A (en) * | 2018-01-19 | 2019-08-01 | 本田技研工業株式会社 | Distance calculation device and vehicle control device |
| CN111516682A (en) * | 2020-05-18 | 2020-08-11 | 合肥杰代机电科技有限公司 | Motor vehicle management and control method, device and system based on intelligent driving environment measurement and control |
| WO2022028772A1 (en) * | 2020-08-06 | 2022-02-10 | Zf Friedrichshafen Ag | Method for operating a steer-by-wire steering system of a motor vehicle, control device, computer programme, and steer-by-wire steering system |
| CN115848402A (en) * | 2022-12-23 | 2023-03-28 | 中汽院智能网联科技有限公司 | Lane line real-time distance measurement system and method based on camera |
-
2023
- 2023-05-22 CN CN202310574408.9A patent/CN116279792B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011128844A (en) * | 2009-12-17 | 2011-06-30 | Fuji Heavy Ind Ltd | Road shape recognition device |
| KR20130005110A (en) * | 2011-07-05 | 2013-01-15 | 현대자동차주식회사 | Method for controlling vehicle interval |
| CN102601862A (en) * | 2012-04-01 | 2012-07-25 | 中联重科股份有限公司 | Engineering vehicle, alerting device and alerting method for safe steering of engineering vehicle |
| JP2019128153A (en) * | 2018-01-19 | 2019-08-01 | 本田技研工業株式会社 | Distance calculation device and vehicle control device |
| CN111516682A (en) * | 2020-05-18 | 2020-08-11 | 合肥杰代机电科技有限公司 | Motor vehicle management and control method, device and system based on intelligent driving environment measurement and control |
| WO2022028772A1 (en) * | 2020-08-06 | 2022-02-10 | Zf Friedrichshafen Ag | Method for operating a steer-by-wire steering system of a motor vehicle, control device, computer programme, and steer-by-wire steering system |
| CN115848402A (en) * | 2022-12-23 | 2023-03-28 | 中汽院智能网联科技有限公司 | Lane line real-time distance measurement system and method based on camera |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116279792B (en) | 2023-07-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108537197B (en) | Lane line detection early warning device and method based on deep learning | |
| Kalinke et al. | A texture-based object detection and an adaptive model-based classification | |
| CN108638999B (en) | Anti-collision early warning system and method based on 360-degree look-around input | |
| CN112987759A (en) | Image processing method, device and equipment based on automatic driving and storage medium | |
| Li et al. | Lane detection based on connection of various feature extraction methods | |
| CN105718872B (en) | Auxiliary method and system for rapidly positioning lanes on two sides and detecting vehicle deflection angle | |
| CN112507862B (en) | Vehicle orientation detection method and system based on multitasking convolutional neural network | |
| CN112800986B (en) | Vehicle-mounted camera external parameter calibration method and device, vehicle-mounted terminal and storage medium | |
| CN107886043B (en) | Vision-aware anti-collision early warning system and method for forward-looking vehicles and pedestrians of automobile | |
| WO2020207528A1 (en) | Method and processing unit for ascertaining the size of an object | |
| CN110532876A (en) | Night mode camera lens pays detection method, system, terminal and the storage medium of object | |
| CN112927303B (en) | Lane line-based automatic driving vehicle-mounted camera pose estimation method and system | |
| CN115082881A (en) | Target detection method, storage medium, electronic device, and vehicle | |
| CN110532875A (en) | Night mode camera lens pays the detection system, terminal and storage medium of object | |
| CN115436920A (en) | Laser radar calibration method and related equipment | |
| US20220171975A1 (en) | Method for Determining a Semantic Free Space | |
| CN116279792B (en) | Vehicle-mounted intelligent steering control system | |
| WO2020169386A1 (en) | Method and computer program product for adapting the resolution of a multisensor system of a road vehicle, and controller for a multisensor system and computer program product | |
| CN115214570B (en) | Vehicle hierarchical braking control method and system combining distance and time factors | |
| CN114359859A (en) | Method and device for processing target object with shielding and storage medium | |
| CN114663860A (en) | Lane line recognition system and method for lane departure system | |
| CN112001344A (en) | Motor vehicle target detection device and method | |
| CN111626180A (en) | Lane line detection method and device based on polarization imaging | |
| DE102019220615A1 (en) | Method and device for recognizing and classifying objects | |
| Bertozzi et al. | Obstacle detection for start-inhibit and low speed driving |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |