Camera-based vehicle adaptive cruise control method

Through the adaptive cruise control method based on camera, the vehicle target and lane line are identified, and the variable workshop time distance and attenuation index function are combined, the problem of poor distance control in traditional systems under different working conditions is solved, and the urban roads is improved Safety and comfort.

CN118991758APending Publication Date: 2024-11-22FAW CAR CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411298337.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-18
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

Traditional adaptive cruise control systems are difficult to accurately control the distance between the vehicle and the vehicle in front under different road conditions, making it difficult to deal with emergency entry conditions on urban roads, which may lead to collision accidents.

Method used

Adaptive cruise control method based on camera is adopted, and target screening strategies and longitudinal control strategies are established by identifying vehicle target objects and lane line information, and variable workshop time distance and attenuation index functions are used to achieve flexible response to different working conditions and Smooth transition.

Benefits of technology

Effectively screen out recommended targets for dangerous areas, improve safety and comfort in complex traffic flows, avoid frequent queue jumps, and ensure driving safety and comfort.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118991758A_ABST
    Figure CN118991758A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of wheel cruise control, and particularly relates to a vehicle self-adaptive cruise control method based on a camera. A target screening strategy is established, recommended targets in a dangerous area are effectively screened out, a camera capable of processing data is used for obtaining front target information and road information, a dangerous target area is defined, and a target closest to a main vehicle in the dangerous targets in the dangerous target area serves as a recommended target vehicle. The position of a target vehicle is analyzed and calculated through camera output information, a longitudinal control strategy of a decision-making layer is established, hierarchical control is achieved, and redundancy and complexity of a control structure are avoided; the actuator control layer receives a final expected acceleration and deceleration instruction from the decision-making layer, and adjusts and controls vehicle power and a braking system by adopting a dynamic feed-forward and PID control algorithm, so that the motion state of the vehicle is controlled, and the self-adaptive cruise of the vehicle is realized; according to the invention, the camera is used to identify the vehicle target object and the lane line information, and a longitudinal control strategy is combined, so that the working condition that the vehicle follows the vehicle and the side vehicle cuts in can be effectively handled.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field of wheel cruise control, and in particular relates to a vehicle adaptive cruise control method based on a camera. Background Art

[0002] The Adaptive Cruise Control (ACC) system is the most common and typical driver assistance system in the automotive industry. It detects the status of the target vehicle in front through the sensors installed in the vehicle, and combines the appropriate vehicle longitudinal control strategy to obtain the following speed and following distance that meet driving safety, thereby realizing functions such as automatic following, cruise control, emergency braking, etc., alleviating the driver's visual and physical fatigue and ensuring driving comfort and safety.

[0003] During the cruise control process, controlling the safe distance between the front vehicle and the front vehicle can improve driving safety and comfort. In the prior art, a fixed vehicle distance or a fixed cruise time interval is usually used as the judgment condition for switching between cruise control and cruise control. This will result in fixed judgment conditions being used to switch modes regardless of the driving conditions of the front vehicle. It is impossible to switch and accurately control the various conditions of the front vehicle, resulting in a stiff switching and poor comfort.

[0004] In traditional ACC high-speed environments, the driving conditions and road environment are relatively simple, the vehicle speed is high, and the vehicle distance is far. In urban road conditions, the traffic flow is relatively dense, the vehicle distance is short, and the vehicle speed is low. The target recommendation method of the traditional ACC system is difficult to cope with the more urgent cutting-in conditions on urban roads, which leads to collision accidents. Summary of the invention

[0005] In order to overcome the above problems, the present invention provides a camera-based vehicle adaptive cruise control method, which utilizes the camera to identify vehicle targets and lane line information, and combines the longitudinal control strategy to effectively deal with the condition where the vehicle cuts in while following a vehicle.

[0006] A camera-based vehicle adaptive cruise control method includes the following contents:

[0007] Step 1: Establish a target screening strategy to effectively screen out recommended targets in dangerous areas;

[0008] Step 1.1, using a camera capable of processing data to obtain information about the target ahead and the road;

[0009] Step 1.2, define the dangerous target area, and take the target closest to the main vehicle in the dangerous target area as the recommended target vehicle;

[0010] Step 1.3, using the camera output information to analyze and calculate the position of the target vehicle;

[0011] Assume that the main vehicle always drives in the center of a fixed lane and does not actively change lanes;

[0012] In straight-line driving conditions: the camera coordinate system is established with the center point of the camera as the origin O, the longitudinal direction of the vehicle as the X-axis, and the lateral direction of the vehicle as the Y-axis. Point P is the point on the target vehicle closest to the main vehicle, and point Q is the point on the main vehicle's driving center trajectory closest to point P. The line segment PQ is the lateral offset distance of the target vehicle relative to the main vehicle's trajectory, expressed as L y express;

[0013] Under the condition of driving on a curve: draw a straight line perpendicular to the X-axis of the camera coordinate system through point P. The intersection of this straight line and the center trajectory of the main vehicle is point R. Point N is the intersection of the tangent of the center trajectory of the main vehicle at point R and the X-axis of the camera coordinate system. L is recorded as the longitudinal distance of the target vehicle, that is, the projection length of the line segment OQ on the X-axis. Angle α is the azimuth of the target vehicle.

[0014] Step 1.4, calculate the lateral offset distance of the target vehicle, where:

[0015] When the vehicle is driving on a straight road, the lateral offset distance L between the target vehicle and the host vehicle y It is expressed as:

[0016] L y = Ltanα(2)

[0017] When the vehicle is driving on a curve,

[0018] Where: the longitudinal distance L of the target vehicle is also the coordinate of the target vehicle on the X-axis of the camera coordinate system, and F(L) is the coordinate of the target vehicle on the Y-axis;

[0019] Step 2: Establish a vertical control strategy for the decision-making layer to achieve hierarchical control and avoid redundancy and complexity of the control structure;

[0020] Step 2.1, taking the variable headway between vehicles as the safety distance model to solve the final expected acceleration, the specific contents are as follows:

[0021]

[0022] Where d safe is the desired safety distance, v host is the speed of the main vehicle, v pre is the target vehicle speed, v rel is the relative speed of the two vehicles, t 0 、c v are all constants greater than zero, t hIndicates the time distance between the cruise control system and the vehicle, d 0 It is the minimum safe distance when parking;

[0023] In order to avoid frequent queue jumping caused by a long following distance, a saturation function is introduced to control the time interval between vehicles within a reasonable range:

[0024]

[0025] Where sat() is the saturation function; t h_max is the upper limit of workshop time interval; t h_max is the lower limit of workshop time interval;

[0026] Assume that the relative distance between the host vehicle and the target vehicle at the initial moment is d act After a period of control, it enters the steady-state following mode; after entering the steady-state following state, the actual distance between the two vehicles is close to the expected safety distance d safe ; According to formula (5), the following formula is obtained:

[0027]

[0028] Δd is the change in the relative position of the two vehicles, and Δv is the relative speed of the two vehicles;

[0029] Derivative of equation (6) gives the relative velocity and relative acceleration

[0030]

[0031] Among them: a h 、a p They are the acceleration of the host vehicle and the acceleration of the target vehicle respectively;

[0032] Based on the above formula, the following state space expression is established:

[0033]

[0034] The above state space equation is transformed into:

[0035]

[0036] Where u represents the control input variable of the system, i.e., the acceleration of the main vehicle; w represents the disturbance of the system, i.e., the acceleration of the target vehicle in front; according to the state equation (9), the optimal control is sought to minimize the following objective performance function:

[0037]

[0038] Where T is the transpose, Q and R are different weight matrices, and t represents time;

[0039] To meet the above requirements, the weight matrices Q and R are defined as:

[0040]

[0041] where q 1 ,q 2 and r are different weight matrix coefficients;

[0042] Solving the target performance function (10) yields the algebraic equation solution u and the feedback gain matrix K, K = [k 1 k 2 ], where k 1 and k 2 are different gain coefficients, i.e., ratio coefficients; thus, it can be concluded that in the steady-state following mode, the expected acceleration a 0 for:

[0043] a 0 =u=-k 1 (d act -v host ·t h -d 0 )-k 2 (v pre -v host ) (12)

[0044] In order to smooth the vehicle acceleration process, a decaying exponential function is introduced, and the improved form is as follows:

[0045]

[0046] where a des is the final expected acceleration, A n represents the attenuation exponential coefficient; T s represents the time history of distance control;

[0047] Step 2.2, adopt the speed cruise control algorithm to solve the final expected acceleration, add the decay exponential function on the basis of the classic PI control algorithm, and smoothly transition from dynamic to steady state. The specific contents are as follows:

[0048]

[0049] In the formula, a 0 is the expected acceleration determined when the vehicle cruises in a steady state, v set Indicates the cruising speed preset by the driver, t 0 ,t f are the start and end time of the integration phase, k p+ / k p- Respectively represent the proportional coefficient when the speed difference is positive / negative; ki+ / k i- Respectively represent the integral coefficient when the speed difference is positive / negative;

[0050] Step 2.3, two control strategies are formed for the two control algorithms of the above formula (13) and formula (15); when Δd>d a When , the system satisfies the speed control following strategy, and the expected acceleration is obtained using formula (13). a When , the system satisfies the distance control following strategy, and the expected acceleration is obtained using formula (15); the distance boundary value d for switching the control strategy is a The calculation formula is as follows:

[0051]

[0052] Step 3: The actuator control layer receives the final desired acceleration and deceleration instructions from the decision layer, and uses the dynamic feedforward plus PID control algorithm to adjust and control the vehicle power and braking system, thereby controlling the vehicle's motion state and realizing the vehicle's adaptive cruise.

[0053] The step 1.1 uses a camera capable of processing data to obtain the target information and road information in front, including the ID of the target, the longitudinal distance of the target, the speed of the target, the left and right deflection angles of the target outline, the width of the target, the lane where the target is located, the motion state of the target, the longitudinal distance of the identifiable lane line, the lane line recognition quality, and the polynomial coefficient C. i .

[0054] Beneficial effects of the present invention:

[0055] The present invention sets a danger zone and divides the danger levels of vehicles within the visual field, thereby better screening out recommended targets and avoiding interference from other running vehicles in the case of complex traffic flows.

[0056] The present invention designs a hierarchical control of the cruise system, separates the decision layer and the actuator control layer, ensures the adaptability of the cruise system to different working conditions, and avoids the complexity of designing the system architecture according to different working conditions.

[0057] The present invention adds exponential functions to the decision layer and the actuator control layer to smoothly transition the changing process of the vehicle driving state, thereby ensuring the safety and comfort of driving. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings used in the description of the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the contents of the embodiments of the present invention and these drawings without creative work.

[0059] Figure 1 It is a flow chart of the implementation steps of the present invention.

[0060] Figure 2 This is the relative position analysis of the target vehicle of the present invention.

[0061] Figure 3 It is a schematic diagram of the dangerous target area of ​​the present invention.

[0062] Figure 4 This is the speed and distance control strategy switching curve of the present invention.

[0063] Figure 5 Schematic diagram of the change of acceleration of the present invention.

[0064] Figure 6 Schematic diagram of vehicle speed change according to the present invention. DETAILED DESCRIPTION

[0065] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0066] In the description of the present invention, unless otherwise clearly specified and limited, the terms "connected", "connected", and "fixed" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0067] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may include that the first and second features are in direct contact, or may include that the first and second features are not in direct contact but are in contact through another feature between them. Moreover, a first feature being "above", "above" and "above" a second feature includes that the first feature is directly above and obliquely above the second feature, or simply indicates that the first feature is higher in level than the second feature. A first feature being "below", "below" and "below" a second feature includes that the first feature is directly below and obliquely below the second feature, or simply indicates that the first feature is lower in level than the second feature.

[0068] In the description of this embodiment, the terms "upper", "lower", "left", "right" and other directions or positional relationships are based on the directions or positional relationships shown in the drawings, and are only for the convenience of description and simplification of operation, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operate in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first" and "second" are only used to distinguish in the description and have no special meaning.

[0069] Example 1

[0070] A camera-based vehicle adaptive cruise control method, such as Figure 1 As shown, including the following:

[0071] Step 1: Establish a target screening strategy to effectively screen out recommended targets in dangerous areas;

[0072] Step 1.1, use a camera that can process data (MOBIEYE 8CONNECT monocular camera with built-in data processing algorithm, ME8 Connect is a monocular ADAS product based on Mobileye's fourth-generation in-vehicle AI chip platform) to obtain the target information and road information in front, including the target ID, the longitudinal distance of the target, the speed of the target, the left and right angles of the target outline, the width of the target, the lane where the target is located, the motion state of the target (including speed information, position and posture information), the longitudinal distance of the identifiable lane line, the lane line recognition quality (the degree of lane line recognition), and the polynomial coefficient C used to fit the lane line i ;

[0073] Step 1.2, define the dangerous target area and determine the recommended target. The dangerous target area is generated based on the actual vehicle road test data and the camera in 1.1 is used to obtain the dangerous area.

[0074] The dangerous target area is to screen out the most suitable system-recommended targets within the camera's field of view under driving conditions with heavy traffic, thereby reducing the occurrence of collision accidents; Figure 3As shown in the figure, the part shown by two blue trapezoids is the dangerous target area, the blue square represents the main vehicle, the green square represents the ordinary target, and the red square represents the potential dangerous target. Among the dangerous targets in the dangerous target area, the target closest to the main vehicle is the recommended target vehicle;

[0075] Step 1.3, using the camera output information to analyze and calculate the position of the target vehicle;

[0076] Based on step 1.2, the target vehicle ID is confirmed and the target object information is output (the ID is used to distinguish different vehicles within the camera's field of view). Assuming that the main vehicle always drives in the center of a fixed lane and does not actively change lanes, the specific situation of the target vehicle position analysis is as follows: Figure 2 As shown; the left and right figures correspond to the two driving conditions of straight line and curve respectively. The white box in the figure represents the main vehicle, and the black box represents the target vehicle;

[0077] In straight-line driving conditions: the camera coordinate system is established with the center point of the camera as the origin O, the longitudinal direction of the vehicle as the X-axis, and the lateral direction of the vehicle as the Y-axis. Point P is the point on the target vehicle closest to the main vehicle, and point Q is the point on the main vehicle's driving center trajectory closest to point P. The line segment PQ is the lateral offset distance of the target vehicle relative to the main vehicle's trajectory, expressed as L y express;

[0078] Under the condition of driving on a curve: draw a straight line perpendicular to the X-axis of the camera coordinate system through point P, and the intersection of the straight line with the center trajectory of the main vehicle is point R, A and B are the horizontal and vertical coordinates of point P respectively; point N is the intersection of the tangent of the center trajectory of the main vehicle at point R and the X-axis of the camera coordinate system; L is recorded as the longitudinal distance of the target vehicle, that is, the projection length of the line segment OQ on the X-axis (point Q is the point closest to point P on the center trajectory of the main vehicle); angle α is the azimuth of the target vehicle (connecting the origin of the camera coordinate system and point P as line segment OP, the angle between line segment OP and the X-axis of the camera coordinate system is angle α); angle β is the angle between line segment QN and the X-axis of the camera coordinate system; angle γ is the angle between line segment PQ and line segment RP;

[0079] Step 1.4, based on the above information, calculate the lateral offset distance of the target vehicle, where the vehicle driving trajectory is expressed as:

[0080]

[0081] When the vehicle is traveling on a straight road, C i =0, (i=0,1,2,3), the lateral offset distance L between the target vehicle and the main vehicle y It is expressed as:

[0082] L y = Ltanα (2)

[0083] When the vehicle is driving on a curve,

[0084] Where: the longitudinal distance L of the target vehicle is also the coordinate of the target vehicle on the X-axis of the camera coordinate system, and F(L) is the coordinate of the target vehicle on the Y-axis;

[0085] Step 1.5: After calculating the lateral offset distance between the host vehicle and the target vehicle according to formula (3), the lateral offset speed V of the host vehicle and the target vehicle is y V y =L′ y , which is used to judge the danger level of the target vehicle. In order to smooth the derivative result, a first-order filter is added, namely V y (n) = aL′ y (n)+(1-a)V y (n-1); where: a is the filter coefficient selected by yourself, and n represents the nth moment;

[0086] Step 2: Establish the vertical control strategy of the decision-making layer and the actuator layer to achieve hierarchical control and avoid the redundancy and complexity of the control structure;

[0087] Decision control layer algorithm design

[0088] The decision-making control layer strategy design includes safety distance model, predictive control algorithm, and improved proportional integral control algorithm.

[0089] Step 2.1, the safety distance model is divided into fixed safety distance and variable safety distance, and the variable safety distance is further divided into constant vehicle headway and variable vehicle headway. This paper takes variable vehicle headway as the safety distance model to solve the final expected acceleration. Under the premise of safety, the safety distance can be appropriately reduced to avoid the situation where the front vehicle cuts in due to excessive following distance. The specific contents are as follows:

[0090]

[0091] Where d safe is the desired safety distance, v host is the speed of the main vehicle, v pre is the target vehicle speed, v rel is the relative speed of the two vehicles, t 0 、c v are all constants greater than zero, t h It represents the time distance between the vehicles of the cruise control system ACC and the vehicle, and is the linear coefficient of the system, d 0 It is the minimum safe distance when parking;

[0092] In order to avoid frequent queue jumping caused by a long following distance, a saturation function is introduced to control the time interval between vehicles within a reasonable range:

[0093]

[0094] Where sat() is the saturation function; t h_max is the upper limit of workshop time interval; t h_max is the lower limit of workshop time interval;

[0095] Assume that the relative distance between the host vehicle and the target vehicle at the initial moment is d act After a period of control, it enters the steady-state following mode; after entering the steady-state following state, the actual distance between the two vehicles is close to the expected safety distance d safe ; According to formula (5), the following formula is obtained:

[0096]

[0097] Δd is the change in relative position of the two vehicles after they travel a certain distance, and Δv is the relative speed of the two vehicles;

[0098] Taking the derivative of equation (6) we can get the relative velocity and relative acceleration

[0099]

[0100] Among them: a h 、a p They are the acceleration of the host vehicle and the acceleration of the target vehicle respectively;

[0101] Based on the above formula, the following state space expression is established:

[0102]

[0103] The above state space equation is transformed into:

[0104]

[0105] Where u represents the control input variable of the system, i.e., the acceleration of the main vehicle; w represents the disturbance of the system, i.e., the acceleration of the target vehicle in front; and x is the state variable of the system. According to the state equation (9), the optimal control is sought to minimize the following objective performance function:

[0106]

[0107] Where T is the transpose, Q and R are different weight matrices. Since the state variables, input variables, and weights can change over time, they are represented by x(t), u(t), Q(t), and R(t), where t represents time. The entire objective function is expressed in a fixed form of optimal control, and the integral is used to obtain the function sum J for a certain time, and this sum is minimized.

[0108] To meet the above requirements, the weight matrices Q and R of the system are defined as:

[0109]

[0110] where q 1 ,q 2 and r are different weight matrix coefficients;

[0111] Solving the target performance function (10) yields the algebraic equation solution u and the feedback gain matrix K, K = [k 1 k 2 ], where k 1 and k 2 are different gain coefficients, i.e., ratio coefficients; thus, it can be concluded that in the steady-state following mode, the expected acceleration a 0 for:

[0112]

[0113] In order to smooth the vehicle acceleration process, a decaying exponential function is introduced, and the improved form is as follows:

[0114]

[0115] where a des is the final expected acceleration, A n represents the attenuation exponential coefficient, which directly determines the time it takes for the system to enter a steady state from a dynamic state, as well as the peak acceleration of the decision; T s represents the time history of distance control;

[0116] Step 2.2, adopt the speed cruise control algorithm to solve the final expected acceleration, add the decay exponential function on the basis of the classic PI control algorithm, and smoothly transition from dynamic to steady state. The specific contents are as follows:

[0117]

[0118] In the formula, a 0 is the expected acceleration determined when the vehicle cruises in a steady state, v host Indicates the speed of the host vehicle, v set Indicates the cruising speed preset by the driver, t 0 ,t f are the start and end time of the integration phase, k p+ / k p- Respectively represent the proportional coefficient when the speed difference is positive / negative; k i+ / k i- Respectively represent the integral coefficient when the speed difference is positive / negative;

[0119] Step 2.3, for the two control algorithms of formula (13) and formula (15), two control strategies are formed (the strategy based on distance in formula (13) and the control strategy based on speed in formula (15)); when Δd>d a When , the system satisfies the speed control following strategy, and the expected acceleration is obtained using formula (13). a When , the system satisfies the distance control following strategy, and the expected acceleration is obtained using formula (15); the distance boundary value d for switching the control strategy is a The calculation formula is as follows:

[0120]

[0121] like Figure 4 As shown, the lower half of the red line represents the distance control following strategy, and the upper half of the red line represents the speed control following strategy. k1 / k2 is the slope of the curve. When the distance and relative speed between the target vehicle and the vehicle satisfy the corresponding functional relationship, the corresponding control strategy is selected.

[0122] Step 3: The actuator control layer receives the final desired acceleration and deceleration instructions from the decision layer, and uses the dynamic feedforward plus PID control algorithm to adjust and control the vehicle power and braking system, thereby controlling the vehicle's motion state and realizing the vehicle's adaptive cruise.

[0123] Example 2

[0124] 1. Target screening strategy

[0125] 1.1 Use the camera to output target information and road information, including the type of target, the ID of the target, the longitudinal distance of the target, the speed of the target relative to the vehicle, the left and right deviation angles of the target outline, the width of the target, the lane where the target is located, the movement state of the target, the state of the target's brake lights, the longitudinal distance of the identifiable lane line, the lane line recognition quality, the polynomial coefficient, the width and type of the lane line.

[0126] 1.2 Use the camera output information to analyze and calculate the position of the target vehicle. Assuming that the vehicle always drives in the center of a fixed lane and does not actively change lanes, the specific situation of the target vehicle position analysis is as follows: Figure 2 As shown. The left and right figures correspond to the two driving conditions of straight line and curve respectively. The white box in the figure represents the vehicle, and the black box represents the target vehicle. Point O is the origin of the camera coordinate system; point P is the point where the target object is closest to the main vehicle; point Q is the point closest to the main vehicle's driving center trajectory line from point P. Line segment PQ is the lateral offset distance of the target vehicle relative to the main vehicle's trajectory. It can be expressed by L yIndicates. Point R is the intersection of the foot of the perpendicular point P and the center track line of the main vehicle. Points A and B are the horizontal and vertical coordinates of point P respectively. Point N is the intersection of the tangent line of the main vehicle track line at point and the X-axis. L is the length of the longitudinal distance OQ of the target vehicle. Angle α is the azimuth of the target vehicle. Angle β is the angle between the tangent line QN at point Q and the horizontal axis. Angle γ is the angle between the line segment PQ and the line segment RP.

[0127] 1.3 Based on the above information, the lateral offset distance of the target vehicle is calculated, and the vehicle's driving trajectory can be expressed as:

[0128]

[0129] When the vehicle is traveling on a straight road, C i =0, the lateral offset distance L between the target vehicle and the main vehicle y It can be expressed as:

[0130] L y = Ltanα(2)

[0131] When the vehicle is driving on a curve,

[0132] 1.4 After calculating the lateral offset distance between the host vehicle and the target vehicle according to formula (3), the lateral offset speed V of the host vehicle and the target vehicle is y It can be expressed as V y =L y ′,

[0133] In order to smooth the derivative result, a first-order filter is added, namely V y (n) = aL y ′(n)+(1-a)V y (n-1).

[0134] 1.5 Define the dangerous target area and determine the recommended target. The dangerous target area is to screen out the most appropriate system recommended target under driving conditions with heavy traffic flow to reduce the occurrence of collision accidents. Figure 3 As shown in the figure, the part shown by two blue trapezoids is the dangerous target area, the blue square represents the vehicle, the green square represents the ordinary target, and the red square represents the potential dangerous target. Among the dangerous targets, the target closest to the vehicle is screened as the dangerous recommended target.

[0135] 2. Vertical control strategy

[0136] 2.1 Decision control layer algorithm design

[0137] The decision-making control layer strategy design includes safety distance model, predictive control algorithm, and improved proportional integral control algorithm.

[0138] The safety distance model is divided into fixed safety distance and variable safety distance. The variable safety distance is further divided into constant vehicle headway and variable vehicle headway. This paper adopts variable vehicle headway. Under the premise of safety, the safety distance can be appropriately reduced to avoid the situation where the front vehicle cuts in due to excessive following distance. The specific formula is as follows:

[0139]

[0140] Where d safe is the desired safety distance, v host is the speed of the main vehicle, v pre is the speed of the preceding vehicle, v rel is the relative speed of the two vehicles, t 0 、c v In order to avoid frequent queue jumping caused by too long a distance from the following vehicle, a saturation function is introduced to control the time distance between vehicles within a reasonable range.

[0141]

[0142] Where sat() is the saturation function; t h_max is the upper limit of workshop time interval; t h_max It is the lower limit of the time headway between workshops.

[0143] Assume that the relative distance d between the host vehicle and the target vehicle at the initial moment is act After a period of control, the vehicle enters the steady-state following mode. After this period of time, the main vehicle displacement S host , front vehicle displacement S pre After entering the steady-state following state, the actual distance between the two vehicles is close to the expected safety distance d safe According to formula (5), the following formula can be obtained:

[0144]

[0145] By taking the derivative of equation (6), we can get the relative velocity and relative acceleration:

[0146]

[0147] Based on the above formula, the following state space expression is established:

[0148]

[0149] The above state space equation is transformed into:

[0150]

[0151] Where t hThe time distance between vehicles that ACC follows is the linear coefficient of the system; u represents the acceleration of the main vehicle, which is the control input variable of the system; and w represents the acceleration of the target vehicle in front, which is the disturbance of the system.

[0152] According to the state equation (9), the optimal control is sought to minimize the following objective performance function.

[0153]

[0154] To meet the above requirements, the weighted matrices Q and R of the system are defined as:

[0155]

[0156] Solving the predictive control equations yields the algebraic equation solution u and the feedback gain matrix K, K = [k 1 k 2 ]. It can be concluded that in the steady-state following mode, the expected acceleration can be expressed as:

[0157] a 0 =u=-k 1 (d act -v host ·t h -d 0 )-k 2 (v pre -v host ) (12)

[0158] In order to smooth the vehicle acceleration process, a decaying exponential function is introduced, and the improved form is as follows:

[0159]

[0160] where a des is the final expected acceleration, A n represents the attenuation exponential coefficient, which directly determines the time it takes for the system to enter a steady state from a dynamic state, as well as the peak acceleration of the decision; T s Represents the time history of distance control.

[0161] The speed cruise control algorithm adds a decay exponential function to the classic PI control algorithm to smoothly transition from dynamic to steady state. The specific form is as follows:

[0162]

[0163] In the formula, a 0 is the expected acceleration determined when the vehicle cruises in a steady state, v host Indicates the speed of the host vehicle, v set Indicates the cruising speed preset by the driver, t 0 ,tf are the start and end time of the integration phase, k p+ / k p- Respectively represent the proportional coefficient when the speed difference is positive / negative. i+ / k i- Respectively represent the integral coefficient when the speed difference is positive / negative, a des is the final expected acceleration.

[0164] The present invention forms two control strategies for the above two control algorithms. The specific switching formula is:

[0165] d a =kΔv+d safe (16)

[0166] Specific as Figure 4 As shown, the lower half of the red broken line represents the distance control strategy, and the upper half of the red broken line represents the speed control strategy.

[0167] The actuator control layer receives the desired acceleration and deceleration instructions from the decision layer, and uses the dynamic feedforward plus PID control algorithm to adjust and control the vehicle power and braking system, thereby controlling the vehicle's motion state and realizing the vehicle's adaptive cruise.

[0168] 3. Use simulation software to build a simulation platform to verify the effectiveness of identification strategies and control algorithms.

[0169] The present invention uses Carsim to simulate vehicle dynamics and combines it with the Simulink control model to perform joint simulation to verify the effectiveness of the algorithm. Figure 5 and Figure 6 .in Figure 5 Represents the change between actual acceleration and expected acceleration. Figure 6 This is the speed change of the self-vehicle and the target vehicle under the condition that the vehicle in the adjacent lane cuts in. As shown by the yellow curve in the figure, the self-vehicle sets the cruising speed to 40km / h. The speed of the target vehicle is shown by the purple curve, which gradually accelerates to 50km / h to overtake the self-vehicle and cut into the lane. The speed change of the self-vehicle is shown by the green curve. When the adjacent vehicle cuts in, the self-vehicle recognizes the recommended dangerous target and slows down until the front vehicle completes the entry and enters the safe area. The self-vehicle cruises normally.

[0170] The preferred embodiments of the present invention are described in detail above in conjunction with the accompanying drawings. However, the protection scope of the present invention is not limited to the specific details in the above embodiments. Within the technical concept of the present invention, any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention. These simple variations all belong to the protection scope of the present invention.

[0171] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present invention will not further describe various possible combinations.

[0172] In addition, various embodiments of the present invention may be arbitrarily combined, and as long as they do not violate the concept of the present invention, they should also be regarded as the contents disclosed by the present invention.

Claims

1. A vehicle adaptive cruise control method based on a camera, characterized in that: It includes the following: Step 1: Establish a target screening strategy to effectively screen out recommended targets in dangerous areas; Step 1.1, using a camera capable of processing data to obtain information about the target ahead and the road; Step 1.2, define the dangerous target area, and take the target closest to the main vehicle in the dangerous target area as the recommended target vehicle; Step 1.3, using the camera output information to analyze and calculate the position of the target vehicle; Assume that the main vehicle always drives in the center of a fixed lane and does not actively change lanes; In straight-line driving conditions: the camera coordinate system is established with the center point of the camera as the origin O, the longitudinal direction of the vehicle as the X-axis, and the lateral direction of the vehicle as the Y-axis. Point P is the point on the target vehicle closest to the main vehicle, and point Q is the point on the main vehicle's driving center trajectory closest to point P. The line segment PQ is the lateral offset distance of the target vehicle relative to the main vehicle's trajectory, expressed as L y express; Under the condition of driving on a curve: draw a straight line perpendicular to the X-axis of the camera coordinate system through point P. The intersection of this straight line and the center trajectory of the main vehicle is point R. Point N is the intersection of the tangent of the center trajectory of the main vehicle at point R and the X-axis of the camera coordinate system. L is recorded as the longitudinal distance of the target vehicle, that is, the projection length of the line segment OQ on the X-axis. Angle α is the azimuth of the target vehicle. Step 1.4, calculate the lateral offset distance of the target vehicle, where: When the vehicle is driving on a straight road, the lateral offset distance L between the target vehicle and the host vehicle y It is expressed as: L y =Ltanα(2) When the vehicle is driving on a curve, Where: the longitudinal distance L of the target vehicle is also the coordinate of the target vehicle on the X-axis of the camera coordinate system, and F(L) is the coordinate of the target vehicle on the Y-axis; Step 2: Establish a vertical control strategy for the decision-making layer to achieve hierarchical control and avoid redundancy and complexity of the control structure; Step 2.1, taking the variable headway between vehicles as the safety distance model to solve the final expected acceleration, the specific contents are as follows: Where d safe is the desired safety distance, v host is the speed of the main vehicle, v pre is the target vehicle speed, v rel is the relative speed of the two vehicles, t0, c v are all constants greater than zero, t h It indicates the time distance between the cruise control system and the vehicle, and d0 is the minimum safe distance when parking; In order to avoid frequent queue jumping caused by a long following distance, a saturation function is introduced to control the time interval between vehicles within a reasonable range: Where sat() is the saturation function; t h_max is the upper limit of workshop time interval; t h_max is the lower limit of workshop time interval; Assume that the relative distance between the host vehicle and the target vehicle at the initial moment is d act After a period of control, it enters the steady-state following mode; after entering the steady-state following state, the actual distance between the two vehicles is close to the expected safety distance d safe ; According to formula (5), the following formula is obtained: Δd is the change in the relative position of the two vehicles, and Δv is the relative speed of the two vehicles; Taking the derivative of equation (6) we can get the relative velocity and relative acceleration Among them: a h 、a p They are the acceleration of the host vehicle and the acceleration of the target vehicle respectively; Based on the above formula, the following state space expression is established: The above state space equation is transformed into: Where u represents the control input variable of the system, i.e., the acceleration of the main vehicle; w represents the disturbance of the system, i.e., the acceleration of the target vehicle in front; according to the state equation (9), the optimal control is sought to minimize the following objective performance function: Where T is the transpose, Q and R are different weight matrices, and t represents time; To meet the above requirements, the weight matrices Q and R are defined as: Among them, q1, q2 and r are different weight matrix coefficients; Solving the target performance function (10) yields the algebraic equation u and the feedback gain matrix K, K = [k1 k2], where k1 and k2 are different gain coefficients, i.e., ratio coefficients. It can be concluded that in the steady-state following mode, the expected acceleration a0 is: a0=u=-k1(d act -v host ·t h -d0)-k2(v pre -v host ) (12) In order to smooth the vehicle acceleration process, a decaying exponential function is introduced, and the improved form is as follows: where a des is the final expected acceleration, A n represents the attenuation exponential coefficient; T s represents the time history of distance control; Step 2.2, adopt the speed cruise control algorithm to solve the final expected acceleration, add the decay exponential function on the basis of the classic PI control algorithm, and smoothly transition from dynamic to steady state. The specific contents are as follows: Where a0 is the expected acceleration determined when the vehicle is cruising in a steady state, v set represents the cruising speed set by the driver, t0, t f are the start and end time of the integration phase, k p+ / k p- Respectively represent the proportional coefficient when the speed difference is positive / negative; k i+ / k i- Respectively represent the integral coefficient when the speed difference is positive / negative; Step 2.3, two control strategies are formed for the two control algorithms of the above formula (13) and formula (15); when Δd>d a When , the system satisfies the speed control following strategy, and the expected acceleration is obtained using formula (13). a When , the system satisfies the distance control following strategy, and the expected acceleration is obtained using formula (15); the distance boundary value d for switching the control strategy is a The calculation formula is as follows: Step 3: The actuator control layer receives the final desired acceleration and deceleration instructions from the decision layer, and uses the dynamic feedforward plus PID control algorithm to adjust and control the vehicle power and braking system, thereby controlling the vehicle's motion state and realizing the vehicle's adaptive cruise.

2. A camera-based vehicle adaptive cruise control method according to claim 1, characterized in that: The step 1.1 uses a camera capable of processing data to obtain the target information and road information in front, including the ID of the target, the longitudinal distance of the target, the speed of the target, the left and right deflection angles of the target outline, the width of the target, the lane where the target is located, the motion state of the target, the longitudinal distance of the identifiable lane line, the lane line recognition quality, and the polynomial coefficient C. i .