Control method and device integrating predictive cruise and lane changing decision-making

Abstract

The present application relates to the technical field of autonomous driving functions, and in particular to a control method and device integrating predictive cruise and lane changing decision-making. The method comprises: receiving the accelerations, speeds and positions of a controlled autonomous driving vehicle and surrounding traffic vehicles at a current moment so as to construct an IDM microscopic following model to predict surrounding traffic vehicle states in a preset future long time domain; on the basis of the surrounding traffic vehicle states in the preset future long time domain, constructing an optimization cost function, so as to plan a lane keeping strategy or lane changing strategy; on the basis of the lane keeping strategy or lane changing strategy, using a quintic polynomial to solve for a reference path; on the basis of the reference path, establishing an optimal control problem cost function for continuous intersection predictive cruise control, so as to solve for the optimal cruise speed within a traveling lane; and sending the reference path and the optimal cruise speed within the traveling lane to the controlled autonomous driving vehicle for control. Thus, the problems that existing lane changing decision-making only considers longitudinal motion and fails to pay attention to a development trend of the surrounding environment, possible dangers and the like are solved.

Description

Control method and device integrating predictive cruise and lane change decision TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic driving functions, in particular to a predictive cruise and lane change decision integrated control method and device based on a vehicle-cloud layered architecture under a cloud control platform. BACKGROUND

[0002] Current single-vehicle intelligent driving vehicles are limited by short sight distance, weak computing power and other shortcomings, and it is difficult to realize higher-level automatic driving. In order to comprehensively improve the safety, smoothness and energy saving of intelligent connected vehicles in the intelligent traffic system, it is necessary to fully exert the advantages of the cloud control platform, such as super-vision, strong computing power, large storage space, and design safe and energy-saving lane change and optimal speed cruise decision, and further tap the application of the cloud control system in the automatic driving scene.

[0003] With the development of vehicle networking, intelligent traffic system (ITS) and intelligent networking technology, real-time information available to vehicles is no longer limited to road information. The driving state of surrounding vehicles and even the real-time state of downstream traffic can be transmitted to the target vehicle through networking technology to ensure green and efficient traffic of vehicles. Predictive cruise control is a control method that uses vehicle navigation systems and electronic maps to obtain front road information (such as slope, curvature, speed limit, etc.) through front traffic flow information and prediction technology, plans future economic speed sequences of vehicles, and coordinates each sub-controller, throttle and brake system by the vehicle controller to realize the adjustment and control of cruise speed, realizes automatic driving of vehicles, and achieves the purpose of saving fuel.

[0004] Compared with the predictive cruise control based on the longitudinal motion of the vehicle, the lane change decision considering the prediction information needs to pay attention to the development trend of the surrounding environment and the possible dangerous situations at all times, which has always been a difficulty in the field of automatic driving technology. It is necessary to design an automatic driving lane change strategy considering the motion trajectory of surrounding vehicles and the interaction between vehicles, and combine the characteristics of the cloud control system, such as traffic all-element digital mapping and global performance optimization, to comprehensively improve the safety and comfort of automatic driving vehicles.

[0005] SUMMARY

[0006] The present application provides a predictive cruise and lane change decision integrated control method and device to solve the problem that the existing lane change decision only considers longitudinal motion and cannot pay attention to the development trend of the surrounding environment and the possible dangerous situations.

[0007] The first aspect embodiment of the application provides a control method integrating predictive cruise and lane-changing decision, comprising the following steps: receiving the acceleration, speed and position of a controlled automatic driving vehicle at a current time and the acceleration, speed and position of surrounding traffic vehicles; constructing an IDM microscopic following model according to the acceleration, speed and position of the controlled automatic driving vehicle and the acceleration, speed and position of the surrounding traffic vehicles, so as to predict the state of the surrounding traffic vehicles in a preset future long time domain by using the IDM microscopic following model; constructing an optimization cost function according to the state of the surrounding traffic vehicles in the preset future long time domain, so as to plan a following strategy or a lane-changing strategy of the controlled automatic driving vehicle by using the optimization cost function; based on the following strategy or the lane-changing strategy, solving a reference path of the controlled automatic driving vehicle by using a quintic polynomial; establishing an optimal control problem cost function of continuous intersection predictive cruise control according to the reference path, so as to solve an optimal cruise speed of a driving lane; and sending the reference path and the optimal cruise speed of the driving lane to the controlled automatic driving vehicle, so as to control the controlled automatic driving vehicle.

[0008] Optionally, the expression of the state of the surrounding traffic vehicles in the preset future long time domain is:

[0009] wherein a id (k) is an expected acceleration of the surrounding traffic vehicle, V e is a desired speed of the surrounding traffic vehicle, a max is a maximum acceleration of the surrounding traffic vehicle, V id (k) is a speed of any surrounding traffic vehicle at time k, Δv id (k) is a longitudinal relative speed of any vehicle and its preceding vehicle at time k, δ a is an acceleration exponent, Δx id (k) is an actual following distance of any vehicle at time k, s * (V id (k), Δv id (k)) is an expected following distance of any vehicle at time k, s0 is a minimum vehicle distance when at rest, THW is a desired vehicle headway, and b is a comfortable deceleration set by the vehicle.

[0010] Optionally, the optimization cost function is constructed according to the state of the surrounding traffic vehicles in the future long time domain, so as to plan a following strategy or a lane-changing strategy by using the optimization cost function, comprising: based on the state of the surrounding traffic vehicles in the future long time domain, generating an optimization cost function combined by a constraint function of the controlled automatic driving vehicle, an obstacle avoidance risk evaluation function and a function of preventing the controlled vehicle from frequently changing lanes; and solving the optimization cost function, and planning the following strategy or the lane-changing strategy of the controlled automatic driving vehicle according to the solving result.

[0011] Optionally, the expression of the optimization cost function is:

[0012] wherein J is the optimization cost function, m is the total mass of the controlled autonomous vehicle, ω1, ω2, ω3 are the weighting coefficients of each optimization function, r1 is the constraint function of the controlled autonomous vehicle, r2 is the obstacle avoidance risk evaluation function, and r3 is the function of preventing the controlled vehicle from frequently changing lanes.

[0013] Optionally, based on the following lane strategy or the lane changing strategy, the reference path is solved by using a quintic polynomial, including: in the case of taking the following lane strategy, selecting the road center line closest to the center coordinates of the controlled autonomous vehicle as the reference path for straight driving in the current lane, and sending the reference path to the controlled autonomous vehicle through the cloud control platform; in the case of taking the lane changing strategy, generating a plurality of generalized lane changing trajectory clusters by using the quintic polynomial, solving the coefficients of the quintic polynomial according to the constraint conditions of the longitudinal and lateral motion of the controlled autonomous vehicle, obtaining the optimal lane changing trajectory as the reference path, and sending the reference path to the controlled autonomous vehicle.

[0014] Optionally, the optimal control problem cost function of the continuous intersection predictive cruise control is established according to the optimal following or lane changing trajectory to solve the optimal cruise speed of the driving lane, including: based on the optimal following or lane changing trajectory, establishing a comprehensive energy consumption function, a driving time cost function, a speed fluctuation cost function, and an overspeed penalty cost function; according to the comprehensive energy consumption function, the driving time cost function, the speed fluctuation cost function, and the overspeed penalty cost function, the optimal control problem cost function of the continuous intersection predictive cruise control is established to solve the optimal cruise speed of the driving lane, and the optimal cruise speed of the driving lane is sent to the controlled autonomous vehicle.

[0015] Optionally, it further includes: constructing a three-degree-of-freedom vehicle dynamics model of the controlled autonomous vehicle in the longitudinal, lateral, and yaw planes; setting a dynamics-constrained MPC controller in the controlled autonomous vehicle; taking the reference path and the optimal cruise speed of the driving lane as the reference quantity of the dynamics-constrained MPC controller, so as to perform dynamics constraint on the controlled autonomous vehicle according to the three-degree-of-freedom vehicle dynamics model in the longitudinal, lateral, and yaw planes and the reference quantity.

[0016] Optionally, the expression of the three-degree-of-freedom vehicle dynamics model in the longitudinal, lateral, and yaw planes is:

[0017] wherein m is the total mass of the controlled autonomous vehicle, is the longitudinal acceleration, is the lateral acceleration, V x is the longitudinal velocity, V y is the lateral velocity, γ is the yaw rate, is the yaw acceleration, I z is the moment of inertia of the vehicle body rotating around the z-axis, F x , F y and M z are the longitudinal force, the lateral force and the total yaw moment of the controlled autonomous vehicle, respectively.

[0018] Optionally, the reference path and the optimal cruise speed of the driving lane are taken as reference quantities of the MPC controller, and dynamics constraints are imposed on the controlled autonomous vehicle according to a three-degree-of-freedom vehicle dynamics model in the longitudinal, lateral and yaw planes and the reference quantities, including: in the case where the controlled autonomous vehicle is executing the following strategy, the reference path and the optimal cruise speed of the driving lane are taken as reference quantities of the MPC controller, an MPC objective function is constructed according to a dynamics model in the longitudinal direction and the reference quantities, so as to impose dynamics constraints on the longitudinal direction of the controlled autonomous vehicle; in the case where the controlled autonomous vehicle is executing the lane-changing strategy, the reference path and the optimal cruise speed of the driving lane are taken as reference quantities of the MPC controller, an MPC objective function is constructed according to dynamics models in the lateral and yaw planes and the reference quantities, so as to impose dynamics constraints on the lateral and yaw planes of the controlled autonomous vehicle.

[0019] The second aspect embodiment of the present application provides a control device integrating predictive cruise and lane-changing decision, including: a receiving module configured to receive acceleration, speed and position of a controlled autonomous vehicle at a current time and acceleration, speed and position of surrounding traffic vehicles; a prediction module configured to construct an IDM micro-following model according to the acceleration, speed and position of the controlled autonomous vehicle and the acceleration, speed and position of the surrounding traffic vehicles, so as to predict states of the surrounding traffic vehicles in a preset future long time domain by using the IDM micro-following model; a planning module configured to construct an optimization cost function according to the states of the surrounding traffic vehicles in the preset future long time domain, so as to plan a following strategy or a lane-changing strategy of the controlled autonomous vehicle by using the optimization cost function; a path solving module configured to solve a reference path of the controlled autonomous vehicle by using a quintic polynomial based on the following strategy or the lane-changing strategy; a speed solving module configured to establish an optimal control problem cost function of continuous intersection predictive cruise control according to the reference path, so as to solve an optimal cruise speed of a driving lane; and a control module configured to send the reference path and the optimal cruise speed of the driving lane to the controlled autonomous vehicle, so as to control the controlled autonomous vehicle.

[0020] The third aspect of the embodiments of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the program to implement the control method integrating predictive cruise and lane-changing decision as described in the above embodiments.

[0021] The fourth aspect of the embodiments of the present application provides a computer program product, which, when executed by a processor, implements the control method integrating predictive cruise and lane-changing decision as described above.

[0022] The fifth aspect of the embodiments of the present application provides a computer readable storage medium, which stores a computer program, and the program, when executed by a processor, implements the control method integrating predictive cruise and lane-changing decision as described above.

[0023] The control method and device integrating predictive cruise and lane-changing decision proposed in the embodiments of the present application propose a vehicle-cloud hierarchical architecture based on the control method integrating predictive cruise and lane-changing decision, which fully takes advantage of the long sight and large amount of calculation of the intelligent networked vehicle cloud control platform, deploys a predictive cruise and safe lane-changing strategy based on prediction information on the cloud, obtains the state of surrounding vehicles in real time through vehicle-road cloud collaborative perception and communication technology, provides technical support and rich information sources for predicting the state of the vehicle, deploys an optimal cruise speed control instruction analysis module, a lane-changing path tracking control module and a control unit on the vehicle, considers the influence of dynamics constraints, actuator output limits and planned paths on the driving safety of the vehicle, and designs a safe-energy-saving model switching strategy based on the vehicle-cloud hierarchical collaborative decision control architecture; the lane-changing decision and optimal cruise speed planning method based on the prediction of surrounding vehicle traffic flow information can improve the driving fluency, comfort and energy saving of the vehicle. The method first uses the designed predictor to obtain the position, speed and acceleration prediction information of surrounding vehicles in a long time domain, and then obtains the optimal lane-changing / following strategy through the artificial potential field method in the lane-changing safety judgment and reference trajectory generation optimizer module according to the prediction information and the current state of the controlled vehicle SV, so as to obtain the corresponding reference path through the quintic function. Subsequently, the optimal cruise speed optimizer is designed to consider the cost function of green wave passage, energy consumption, driving time, speed fluctuation and overspeed penalty, so as to obtain the optimal cruise speed of the driving lane based on the idea of dynamic programming inverse solution. An MPC controller considering cloud decision is designed on the vehicle, which is based on the dynamics model of the controlled vehicle SV, takes the optimal cruise speed and reference path transmitted by the cloud control platform as the reference quantity of the MPC controller, and constrains the state quantity, control quantity and control quantity change rate of the state space equation of the MPC controller, so as to comprehensively improve the path tracking performance and stability of the normal system.

[0024] Additional aspects and advantages of the present application will be made apparent from the following description with reference to the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0025] The above and / or additional aspects and advantages of the present application will become apparent and be readily appreciated from the following description, including the accompanying drawings, wherein:

[0026] Fig. 1 is a specific operation scenario diagram of a vehicle driving on a highway according to an embodiment of the present application;

[0027] Fig. 2 is a specific execution framework diagram of a control method of integrating predictive cruise and lane change decision according to an embodiment of the present application;

[0028] Fig. 3 is a flowchart of a control method of integrating predictive cruise and lane change decision according to an embodiment of the present application;

[0029] Fig. 4 is a scenario diagram of a vehicle driving on a highway according to an embodiment of the present application;

[0030] Fig. 5 is a path generation schematic diagram according to an embodiment of the present application;

[0031] Fig. 6 is a path planning switching strategy diagram according to an embodiment of the present application;

[0032] Fig. 7 is a block schematic diagram of a control device of integrating predictive cruise and lane change decision according to an embodiment of the present application;

[0033] Fig. 8 is a structural schematic diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0034] Embodiments of the present application are described in detail below with reference to the accompanying drawings, examples of which are shown in the drawings, wherein the same or similar notations represent the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to explain the present application, and cannot be understood as limiting the present application.

[0035] The predictive cruise and lane change decision integrated control method and device of the embodiments of the present application are described below with reference to the accompanying drawings. In view of the problem that the existing lane change decision mentioned in the above background art only considers longitudinal motion and cannot pay attention to the development trend of the surrounding environment and the possible dangerous situation, the present application proposes a car-cloud hierarchical architecture, which can effectively solve the problem of limited vehicle space and limited cost of on-board installation of the single vehicle system, so that the vehicle end only needs to focus on the longitudinal and lateral dynamics control of the chassis, and the function modules of decision planning are deployed in the cloud, using the high-performance computing and big data processing capabilities of the cloud to guide the vehicle end to realize long-term predictive cruise and lane change decision. The upper cloud end is responsible for collecting dynamic traffic information on the target traffic path of the service vehicle, providing predictive future traffic flow information for the vehicle end, predicting the trend change of the intersection traffic flow in the future long-term, and planning a globally optimal driving speed trajectory for the single vehicle intelligent driving of the vehicle end. However, the driving range of the vehicle is only on one lane, which will seriously limit the optimization space of the vehicle driving, and it is difficult to adapt to complex traffic conditions by only considering static or regular traffic information. Many existing lane change decision methods are based on the traffic environment currently faced by the vehicle to determine whether to change lanes. At the moment when these methods generate lane change decisions, the vehicle has already lost part of the driving performance or has been in a less ideal driving environment.

[0036] Therefore, the lane change behavior considering the future dynamic prediction information of the surrounding traffic vehicles in the present application can provide more possibilities for optimizing the driving strategy of the controlled vehicle, thereby further improving the comprehensive performance of the vehicle. In addition, at the execution level of the vehicle lane change driving, the path planning stage is designed to achieve the multi-objective of the intelligent traffic system, especially emphasizing the safety, driving comfort and travel efficiency of the vehicle. For this purpose, the present application also designs a lane change path multi-objective optimization function considering the comfort and efficiency of the lane change process, solves the optimal lane change path through a quintic path planning method, and constructs a straight driving and lane change driving path planning switching strategy to finally determine the optimal reference trajectory. In addition, in order to ensure the feasibility of the planned path and meet the vehicle dynamics constraints, a model predictive controller associated with the motion control module is integrated in the path planning stage, so as to ensure that the vehicle end can effectively execute the optimal driving strategy.

[0037] As shown in FIG. 1, in the embodiment of the present application, the optimal vehicle driving strategy based on the future traffic flow information of surrounding vehicles, vehicle model and scene pre-analysis module is deployed in the cloud. The vehicle end proposes a vehicle chassis longitudinal and lateral integrated control method considering the dynamic constraint, actuator output saturation and path tracking performance. The safety lane changing mechanism is designed to meet the demand of vehicle driving safety, and the optimal driving strategy is iteratively updated according to the real-time state of the vehicle at multiple planning points, thereby further improving the driving efficiency and comfort of the vehicle. The cloud control platform is based on the traffic flow information uploaded by the road test sensor, surrounding intelligent networked vehicles and signal lights, and combines the IDM microscopic following model to predict the motion state of the traffic vehicles in the perception range in the future period of time, to solve the expected cruising speed and lane changing decision of the controlled unmanned vehicle, and to send the optimal vehicle speed sequence based on the long time domain and the reference path to the vehicle end. Thus, the problems that the existing lane changing decision only considers the longitudinal motion and cannot pay attention to the development trend of the surrounding environment and the dangerous conditions that may be faced are solved.

[0038] It should be noted that, as shown in FIG. 2, the embodiment of the present application mainly includes two parts, which are the predictive cruise and lane changing decision integrated strategy based on the future traffic vehicle state prediction in the cloud control platform and the vehicle chassis longitudinal and lateral motion integrated control algorithm considering the vehicle chassis in the vehicle end platform. In the predictive cruise and lane changing decision integrated strategy based on the future traffic vehicle state prediction in the cloud control platform, the predictor, the optimal cruising speed optimizer and the lane changing safety judgment and reference trajectory generation optimizer are included. The vehicle chassis longitudinal and lateral motion integrated control algorithm is designed in the vehicle end platform, including the safety execution judgment strategy and the chassis longitudinal and lateral integrated model predictive controller.

[0039] Specifically, FIG. 3 is a flowchart of a predictive cruise and lane changing decision integrated control method provided by the embodiment of the present application.

[0040] As shown in FIG. 3, the predictive cruise and lane changing decision integrated control method includes the following steps:

[0041] In step S301, the acceleration, speed and position of the controlled automatic driving vehicle at the current time and the acceleration, speed and position of the surrounding traffic vehicles are received.

[0042] In step S302, the IDM microscopic following model is constructed according to the acceleration, speed and position of the controlled automatic driving vehicle and the acceleration, speed and position of the surrounding traffic vehicles, so as to predict the state of the surrounding traffic vehicles in the preset future long time domain by using the IDM microscopic following model.

[0043] As shown in FIG. 4, in actual execution, all traffic flow vehicles in the double-lane are defined, the left-lane vehicles are numbered from front to back as l.i, i∈(1, 2, 3…p), the right-lane vehicles are numbered from front to back as r.i, i∈(1, 2, 3…q), and p and q respectively represent all traffic flow vehicles in the left and right lanes. The set ξ id (k) represents the state of the surrounding vehicle at time k, and is represented as:

[0044] wherein x id (k) represents the longitudinal position of the vehicle at time k, V id (k) represents the vehicle speed at time k, lane id (k) represents the current lane number of the vehicle at time k, lane id ={0, 1}, when the vehicle is in the left lane, lane id =1, and when the vehicle is in the right lane, lane id =0.

[0045] The control amount of the surrounding vehicle at time k is represented by the set U id (k):

[0046] wherein a id (k) represents the acceleration of the surrounding vehicle at time k, θ l id (k) represents the left-lane changing decision of the surrounding vehicle at time k, when θ l id (k) is equal to 0, it represents no lane changing, and when θ l id (k) is equal to 1, it represents left-lane changing; and θ r id (k) represents the right-lane changing decision, and the value is taken in the same way as the left-lane changing. The embodiment of the application assumes that the vehicle driving is relatively stable in the highway driving scene, and the surrounding vehicle will not change lanes within the long prediction time domain designed by the cloud control platform, so in the traffic vehicle state prediction method, the lane changing control amount of the traffic vehicle is:

[0047] The IDM microscopic following model is used to calculate the expected acceleration a id (k) of the traffic vehicle, wherein the expression of the IDM microscopic following model is:

[0048] wherein V e represents the expected speed of the surrounding vehicle, a max is the maximum acceleration of the surrounding vehicle, V id (k) represents the speed of any surrounding vehicle at time k, and ΔVid (k) represents the longitudinal relative speed of the arbitrary vehicle and its front vehicle at time k, δ a represents the acceleration index, usually δ a = 4, Δx id (k) represents the actual following distance of the arbitrary vehicle at time k, s * (V id (k), Δv id (k)) represents the expected following distance of the arbitrary vehicle at time k, s0 represents the minimum vehicle distance when stationary, THW represents the expected vehicle headway, and b represents the comfortable deceleration set by the vehicle. Based on the IDM microscopic following model established above, the speed V id (m) of the surrounding vehicle m at time k, the longitudinal position x id (m), and the lane lane id (m) can be iteratively obtained based on its value at time k: V id (m) = V id (k) + a id (k) ΔT + a id (k+1) ΔT…a id (m-1) ΔT

[0049] In a real scene, the error of the prediction result increases as the prediction time domain increases, so the rolling update prediction method is used in the embodiments of the application to eliminate the prediction state error. By continuously updating the current state of the traffic vehicle received by the cloud, the prediction model can continuously update the prediction state of the traffic vehicle, and obtain the prediction information of the surrounding vehicle at time m (i.e., the state of the surrounding traffic vehicle in a preset future long time domain).

[0050] In step S303, an optimization cost function is constructed according to the state of the surrounding traffic vehicle in the future long time domain, so as to plan a following strategy or a lane changing strategy of the controlled autonomous vehicle by using the optimization cost function.

[0051] In some embodiments, the optimization cost function is constructed according to the state of the surrounding traffic vehicle in the future long time domain, so as to plan a following strategy or a lane changing strategy by using the optimization cost function, comprising:

[0052] Based on the state of the surrounding traffic vehicle in the future long time domain, an optimization cost function is generated by combining a constraint function of the controlled autonomous vehicle, an obstacle avoidance risk evaluation function, and a function of preventing the controlled vehicle from frequently changing lanes;

[0053] The optimization cost function is solved, a following strategy or a lane changing strategy is planned according to the solving result, and the following strategy or the lane changing strategy is sent to the controlled autonomous vehicle.

[0054] In actual execution, based on the prediction information of the surrounding vehicles at the m moment, an optimization cost function is designed by combining the controlled vehicle driving efficiency constraint function r1, the obstacle avoidance risk evaluation function r2 and the function r3 of preventing the controlled vehicle from frequently changing lanes, which can be designed as: Wherein, ω1, ω2 and ω3 are the weighting coefficients of each optimization function, and the controlled vehicle speed constraint function r1 can be designed as: Wherein, V max and V min are the highest and lowest vehicle speeds designed for the working condition, V sv represents the current longitudinal speed of the controlled vehicle, and the speed range is kept in V sv ∈ [V min , V max ], and the driving efficiency constraint function r1 designed based on the above can judge whether the speed of the controlled autonomous vehicle SV at the current moment can meet the driving efficiency of safe lane changing.

[0055] The obstacle avoidance risk evaluation function r2 is to ensure the obstacle avoidance ability of the designed cost function under the city light control road, and the artificial potential field function is introduced in the cost function r2, which means that there is no collision with the front surrounding vehicle or the rear surrounding vehicle in the long time domain predicted by the cloud control platform, which can be designed as:

[0056] In addition, considering that frequent lane changing will cause traffic congestion and seriously affect driving safety, the embodiments of the present application use a Gaussian function to describe the lane changing behavior.

[0057] Wherein, X n and Y n represent the longitudinal and transverse coordinates of the surrounding vehicles respectively, σ1, σ2 and σ3 respectively determine the potential risk of collision of the controlled vehicle SV, Y mid is the transverse coordinate of the lane center line, and P lane represents the maximum value of the deviation of the controlled vehicle from the lane center line, when this value is selected small enough, the controlled vehicle can more easily change lanes. Based on the above designed optimization cost function, it can be determined whether the lane changing strategy of the controlled vehicle is following or changing lanes.

[0058] In step S304, based on the following strategy or lane changing strategy, the reference path of the controlled autonomous vehicle is solved by using a quintic polynomial.

[0059] In some embodiments, based on the following strategy or lane changing strategy, the reference path is solved by using a quintic polynomial, including:

[0060] In the case of taking the following strategy, the road center line closest to the center coordinate of the controlled automatic driving vehicle is selected as the reference path for straight driving in the current lane, and the reference path is sent to the controlled automatic driving vehicle through the cloud control platform;

[0061] In the case of taking the lane changing strategy, a plurality of generalized lane changing trajectory clusters are generated by using a quintic polynomial, the coefficients of the quintic polynomial are solved according to the preset constraints of the longitudinal and lateral motion of the controlled automatic driving vehicle, an optimal lane changing trajectory is obtained as the reference path, and the reference path is sent to the controlled automatic driving vehicle.

[0062] In actual execution, if the cloud control platform calculates to take the following strategy, the cloud control platform issues a straight path planning signal to the controlled automatic driving vehicle, the straight path selects the road center line closest to the center coordinate of the vehicle as the reference path for straight driving in the current lane, and the optimal reference vehicle speed is obtained through the speed planner;

[0063] If the cloud control platform calculates to take the lane changing strategy, the cloud control platform will plan a generalized lane changing trajectory of the controlled automatic driving vehicle on a two-lane structured road in real time, design a trajectory cluster model without considering any vehicle dynamics and environmental geometric constraints, and the trajectory cluster model is established by a curve difference method, as shown in FIG. 5, the lane changing process of the controlled automatic driving vehicle SV is divided into three stages: preparation for lane changing stage, formal lane changing stage and end of lane changing stage. The cloud control platform will calculate the optimal lane changing trajectory in the preparation for lane changing stage and send it to the vehicle end; the initial time of the formal lane changing is τ0, the terminal time is τ s , the longitudinal distance of the lane changing trajectory is L s , and the lateral distance is D s ; in the end of lane changing stage, the vehicle enters a new lane, and a new following or lane changing instruction is issued by the cloud.

[0064] Based on the quintic polynomial, a series of unconstrained generalized lane changing trajectory clusters are generated, the position and speed of the controlled automatic driving vehicle SV at the initial time and the terminal time are known, the expressions of the acceleration about time are known, the boundary conditions are substituted to solve the coefficients of the quintic polynomial, and thus the optimal lane changing trajectory is obtained. According to the Pontryagin Minimum Principle (PMP), the optimal boundary value problem is solved, and a cost function based on minimizing the longitudinal and lateral acceleration of the vehicle is constructed:

[0065] At the same time, the constraints of the longitudinal and lateral motion of the vehicle need to be defined:

[0066] wherein D0 and D s are the lateral displacements at the initial and terminal times of lane changing, L0 and L s are the longitudinal displacements at the initial and terminal times of lane changing, Vx (τ0) and V x (τ s ) represent the longitudinal vehicle speed at the beginning and end of the lane change, V y (τ0) and V y (τ s ) represent the lateral vehicle speed at the beginning and end of the lane change, a x (τ0) and a x (τ s ) are the longitudinal acceleration at the beginning and end of the lane change, a y (τ0) and a y (τ s ) are the lateral acceleration at the beginning and end of the lane change.

[0067] The above formula is converted into a Hamilton function form, which can be obtained: H(x(t),u(t),κ(t))=g(x(t),u(t))+κ T f(x(t),u(t)) κ T =(κ1,κ2,κ3) T

[0068] Wherein, f(x(t),u(t)) represents the system equation, g(x(t),u(t)) is the cost function, κ1, κ2 and κ3 represent the Lagrange operator corresponding to the state quantity of Hamilton system equation respectively. According to the Pontryagin minimum principle, through integration, we can get:

[0069] The optimal control input is:

[0070] In order to simplify the calculation, the origin of the earth coordinate system coincides with the geometric center point of the vehicle at the initial time of the formal lane change, the longitudinal and lateral acceleration at the initial time is 0, the longitudinal and lateral acceleration at the end time is 0, and the lateral speed at the initial time and the end time is 0, so the expressions of the displacement, speed, acceleration and jerk of the vehicle in the longitudinal and lateral directions can be obtained respectively as follows:

[0071] The lane change trajectory expression is calculated as a quintic polynomial lane change trajectory model which is the optimal longitudinal and lateral jerk index. Among them, the trajectory model is related to the lane change lateral displacement, longitudinal displacement, initial speed and lane change time, while the lateral displacement D sGenerally determined by the lane width, according to the national highway engineering technical specification standard JTGB01-2014, the lane width of the embodiment of the application is 3.75 m; and the lane-changing longitudinal displacement is related to the vehicle speed and the lane-changing time, and the longitudinal acceleration in the geodetic coordinate system in the embodiment of the application is 0, and the target of minimizing the longitudinal jerk of the vehicle is achieved by keeping the longitudinal acceleration unchanged, so the longitudinal vehicle speed remains unchanged, and the following relationship exists: V x (t) = V x0

[0072] When the vehicle travels at a constant speed, the product of the constant speed and the lane-changing time is equal to the lane-changing longitudinal displacement value, thereby ensuring the comfort during the lane-changing process, and therefore, the lane-changing trajectory expression based on the quintic polynomial after simplification can be finally obtained by combining the above formula as:

[0073] The maximum lateral acceleration a ymax and the maximum lateral jerk jerk ymax in the lane-changing process of the vehicle are selected as indexes for measuring the lane-changing comfort. f The lane-changing efficiency of the vehicle is measured, and the optimal lane-changing time t f is finally solved. ymax The optimal lane-changing path is determined. The normalization processing is performed on each variable, and the optimization function form can be designed as: s.t|a yres |≤a ymax ,|jerk yres |≤jerk fmin ,t f ≤t fmax

[0074] wherein a ymax represents the maximum lateral acceleration constraint allowed by the vehicle, jerk yres represents the maximum lateral jerk constraint allowed by the vehicle, t fmin and t fmax represent the lower limit and the upper limit of the lane-changing duration of the vehicle, ω1 and ω2 represent the weight coefficients of each index, and the weight coefficients satisfy the relationship ω1+ω2=1. The final lane-changing strategy is shown in FIG. 6.

[0075] In step S305, the optimal control problem cost function of the continuous intersection predictive cruise control is established according to the reference path, so as to solve the optimal cruise speed of the driving lane.

[0076] In some embodiments, the optimal control problem cost function of the continuous intersection predictive cruise control is established according to the optimal following or lane-changing trajectory, so as to solve the optimal cruise speed of the driving lane, including:

[0077] Based on the optimal following or lane-changing trajectory, an integrated energy consumption function, travel time cost function, speed fluctuation cost function, and overspeed penalty cost function are established;

[0078] According to the integrated energy consumption function, travel time cost function, speed fluctuation cost function, and overspeed penalty cost function, an optimal control problem cost function of the continuous intersection predictive cruise control is established to solve the optimal cruise speed of the travel lane, and the optimal cruise speed of the travel lane is sent to the controlled automatic driving vehicle.

[0079] In actual execution, when the controlled vehicle completes the following strategy or lane-changing strategy, the speed prediction optimizer establishes an optimal control problem of optimal economic travel speed planning under the consideration of multi-signalized intersection according to the current lane traffic, surrounding traffic state (position, speed, etc.), and travel economy composition target function, and takes the longitudinal dynamics constraint, controlled object target speed interval, and road speed limit as constraint conditions, and the specific process is as follows:

[0080] Firstly, based on the fact that the red light time constraint of the urban road working condition is closely related to the position, and the optimal travel speed needs to be combined with the position information and road information of the controlled vehicle, the time domain problem of the cost function needs to be converted into the distance domain problem, and the conversion formula is:

[0081] Combined with the longitudinal dynamics equation of the controlled automatic driving vehicle SV, the state transition matrix taking the travel time and speed as the state quantity and the motor output torque as the control quantity can be obtained:

[0082] Subsequently, the energy consumption and travel time are established as the main target function, so as to suppress the influence of excessive speed fluctuation on the travel road working condition, and to punish the speed points exceeding the target travel speed interval. The energy consumed by each step of discrete distance point can be obtained through the established energy consumption model, and the energy consumption function J1 considering the discrete distance domain can be designed as:

[0083] Wherein, Q(s h ) represents the energy Q consumed by the travel distance, and the speed change V(h-1) to V(h) in each segment of discrete distance step is a uniform speed or uniform variable speed process, so the travel time t d (or t b ) consumed in the discrete distance step length can be easily obtained from the relationship of Δs and V(h-1), V(h). Wherein, Δs is the discrete distance step length, h represents the stage, h=1, 2, …H. H represents the total number of stages, which is calculated from the road length L trip and the discrete distance step length Δs:

[0084] where L is the length of the route trip is the cumulative sum of the length of the l segments: L trip = L1+ L2+…L i

[0085] where L i is the length of the i-th segment, i = 1, 2Ll, and l represents the total number of segments.

[0086] In order to achieve the multi-objective optimization comprehensive promotion of vehicle driving, the maximum driving economy cannot be completely sacrificed at the cost of travel time. Therefore, smaller travel time needs to be considered in the cost calculation process to ensure the traffic efficiency of the vehicle as much as possible under the same distance domain, and the driving time cost J2 calculation formula is:

[0087] where t trip represents the driving time, and V(h) is the final speed of the road in the h-th stage.

[0088] The speed fluctuation will seriously affect the comfort of the passengers in the vehicle, and the speed fluctuation will also affect the economy of the vehicle driving. Therefore, the cost function J3 considering the speed fluctuation needs to be designed to curb the excessive speed fluctuation to affect the normal driving of the vehicle: J3 = AV(s h ) = |V(h) - V(h-1)|

[0089] The upper and lower bounds of the driving speed of the controlled vehicle are defined, and the speed of the controlled vehicle is kept within the interval , then the average speed can be expressed as follows:

[0090] A cost function that increases the penalty cost when the speed interval is exceeded is designed, which can be expressed as:

[0091] Combining the four optimization cost functions established above, the performance of energy consumption, driving time, speed fluctuation, and overspeed penalty term is integrated, and the optimal control problem cost function of the continuous intersection predictive cruise control is established:

[0092] where s h represents the h-th distance step, is the distance from the starting point and the end point of the h-th stage to the initial position of the vehicle, then λ1, λ2, λ3, λ4 are the weight factors corresponding to the cost items.

[0093] Then, the optimal cruise speed of the driving lane is solved according to the optimal control problem cost function of the continuous intersection predictive cruise control.

[0094] In step S206, the reference path and the optimal cruising speed of the travel lane are sent to the controlled automatic driving vehicle to control the controlled automatic driving vehicle.

[0095] In some embodiments, the embodiments of the present application embed an embedded path tracking controller in the vehicle end to realize speed and path tracking of the controlled vehicle, specifically including the following steps:

[0096] A three-degree-of-freedom vehicle dynamics model of the controlled automatic driving vehicle in the longitudinal, lateral and yaw planes is constructed;

[0097] A dynamics constraint MPC controller is set in the controlled automatic driving vehicle;

[0098] The reference path and the optimal cruising speed of the travel lane are taken as reference quantities of the dynamics constraint MPC controller, and the controlled automatic driving vehicle is dynamically constrained according to the three-degree-of-freedom vehicle dynamics model in the longitudinal, lateral and yaw planes and the reference quantities.

[0099] In some embodiments, the reference path and the optimal cruising speed of the travel lane are taken as reference quantities of the MPC controller, and the controlled automatic driving vehicle is dynamically constrained according to the three-degree-of-freedom vehicle dynamics model in the longitudinal, lateral and yaw planes and the reference quantities, including:

[0100] In the case that the controlled automatic driving vehicle executes a following strategy, the reference path and the optimal cruising speed of the travel lane are taken as reference quantities of the MPC controller, and an MPC objective function is constructed according to the dynamics model in the longitudinal direction and the reference quantities to dynamically constrain the longitudinal direction of the controlled automatic driving vehicle;

[0101] In the case that the controlled automatic driving vehicle executes a lane changing strategy, the reference path and the optimal cruising speed of the travel lane are taken as reference quantities of the MPC controller, and an MPC objective function is constructed according to the dynamics model in the lateral and yaw planes and the reference quantities to dynamically constrain the lateral and yaw planes of the controlled automatic driving vehicle.

[0102] In actual execution, a three-degree-of-freedom vehicle dynamics model of the controlled automatic driving vehicle SV in the longitudinal, lateral and yaw planes and a state space equation thereof are established:

[0103] where m is the total mass of the vehicle, is the longitudinal acceleration, is the lateral acceleration, V x is the longitudinal velocity, V y is the lateral velocity, γ is the yaw angular velocity, is the yaw angular acceleration, I z is the moment of inertia of the vehicle body rotating around the z axis, Fx , F y , and M z are the vehicle longitudinal resultant force, lateral resultant force, and total yaw moment, respectively, and cosδ f ≈ 1, sinδ f ≈ 0, F x , F y M z can be expressed as:

[0104] where a and b are the distances from the vehicle center of gravity CG to the front and rear axles, respectively; l s is the width of the vehicle body, and subscripts lf, lr, rl, and rr represent the vehicle front left, front right, rear left, and rear right, respectively, the vehicle longitudinal force F yf and F yr are:

[0105] where C f and C r are the vehicle front and rear cornering stiffness, respectively, δ f is the vehicle front wheel steering angle, and α f and α r are the vehicle front and rear center of mass cornering angles. The vehicle position in the global coordinate system can be expressed as:

[0106] X and Y are the vehicle longitudinal and lateral global coordinates, respectively, a x is the vehicle longitudinal acceleration, and the state space equation can be obtained by integrating the above equations:

[0107] where the state variable and the input variable u = [a x δ f ] T .

[0108] To facilitate the design of the dynamic constraint MPC controller, the state space equation based on the three-degree-of-freedom vehicle dynamics model is converted into a discrete form with a sampling time of Δτ:

[0109] where

[0110] When the controlled autonomous vehicle SV is performing the following strategy, the controlled autonomous vehicle SV will receive the reference speed transmitted by the cloud control platform as the reference quantity of the dynamics constraint MPC controller, and the reference displacement and yaw angle are both 0. At the same time, in order to avoid the collision between the controlled vehicle and the front vehicle, the safety space constraint (i.e. the longitudinal dynamics model) is integrated into the standard MPC optimization problem, and the objective function of the MPC is expressed as follows:

[0111] Where the designed constraints are:

[0112] Where X pre represents the longitudinal position of the front vehicle, which indicates that the controlled autonomous vehicle SV can track the optimal cruise speed under the condition of safety, P1 and R1 are the weighting coefficients of the cost function, N p is the prediction horizon, and the reference acceleration a x,ref will be the optimization objective of the dynamics constraint MPC controller.

[0113] In addition, the weighting coefficients will also have a great influence on the tracking performance of the controller. In order to better track the reference acceleration, the proportion of the weighting coefficient will be set as: R1 = 4P1, and the prediction horizon is set as N p = 15 to ensure the multi-objective optimization of the dynamics constraint MPC controller.

[0114] When the vehicle needs to perform the lane changing strategy, the controlled autonomous vehicle SV will track the reference path generated by the path planning module. Since the longitudinal speed control is realized by using a separate controller, the standard lateral dynamics model can be used to derive the cost function considering path tracking. In the path planning process, the boundary of the lane changing path also needs to consider the potential collision risk with the surrounding environment vehicles. Therefore, the lateral stability of the controlled autonomous vehicle SV (i.e. the lateral and yaw plane dynamics model) will be designed as a constraint in the dynamics constraint MPC controller, and its objective function can be expressed as:

[0115] Where the following constraints need to be satisfied:

[0116] Where, P2 and R2 are the weighting coefficients of the lane changing cost function, respectively. P2 = diag(P 21 ,P 22 ). The vehicle reference yaw angle Thus, the controlled autonomous vehicle SV always ensures that the heading angle is zero during the lane changing process, and in addition, the optimization function based on the lane changing strategy also needs to consider the path tracking performance, stability performance and steering effective work of the controlled vehicle, therefore, in the present application, in order to preferentially improve the path tracking performance of the dynamics-constrained MPC controller, the weighting coefficient is set as: P 22 = 10P 21 .

[0117] In order to facilitate the design of the dynamics-constrained MPC controller, it is necessary to convert the state constraint into an input constraint. Therefore, the present application designs a standard example to describe this process, that is: -ηξ' max ≤Fξ′0+GV(k)≤ηξ' max

[0118] Wherein, ξ'0 represents the initial state of the controlled autonomous vehicle SV at each sampling time, and the parameter matrix of the dynamics-constrained MPC controller is designed as follows:

[0119] By converting the above established input constraint into an inequality, the state constraint can be easily considered in the design of the dynamics-constrained MPC controller.

[0120] The predictive cruise and lane changing decision integrated control method according to the present application has the following beneficial effects:

[0121] (1) Combining the advantages of the cloud platform and the vehicle platform, based on the wide-range perception ability and powerful computing power advantage of the cloud control system, combined with rich traffic prediction information, a more advanced lane changing strategy and optimal cruise speed planning is obtained, the vehicle platform can obtain a controller algorithm with higher real-time performance and stronger stability by means of the decision planning of the cloud platform, and execute the vehicle acceleration / deceleration and front wheel steering angle instructions for comprehensive improvement of driving economy and safety;

[0122] (2) The predictive cruise and lane changing decision based on surrounding vehicle traffic prediction information is designed, first, the IDM model is used to predict the traffic information of the surrounding vehicles in a long time domain, and the corresponding following / lane changing strategy is obtained by using the artificial potential field method, which fully utilizes the super-vision advantage of the cloud platform, and effectively improves the application scene of the traditional predictive cruise speed; Subsequently, the optimal cruise speed optimizer will obtain the economic driving speed of the lane according to the comprehensive cost optimization function; Compared with the single-lane driving limitation of the traditional predictive cruise speed planning, the predictive cruise control strategy without considering lane changing improves the traffic efficiency of the controlled vehicle, and also considers the economic driving of the controlled vehicle after lane changing;

[0123] (3) receiving the planning vehicle speed and the lane-following / lane-changing strategy transmitted by the cloud on the vehicle terminal platform, designing a vehicle chassis longitudinal and lateral integrated model predictive controller considering the obstacle avoidance risk, the controlled vehicle state quantity and the input quantity constraint; in order to improve the path tracking, speed tracking and obstacle avoidance performance of the designed controller, the weighting coefficients of the controller are optimized, so as to realize the driving economy, smoothness and safety of the controlled vehicle on the multi-signal control urban road.

[0124] Secondly, the predictive cruise and lane-changing decision integrated control device according to the embodiment of the application is described with reference to the accompanying drawings.

[0125] FIG. 7 is a block schematic diagram of the predictive cruise and lane-changing decision integrated control device according to the embodiment of the application.

[0126] As shown in FIG. 7, the predictive cruise and lane-changing decision integrated control device 70 comprises a receiving module 701, a prediction module 702, a planning module 703, a path solving module 704, a vehicle speed solving module 705 and a control module 706.

[0127] The receiving module 701 is configured to receive the acceleration, speed and position of the controlled automatic driving vehicle at the current time and the acceleration, speed and position of the surrounding traffic vehicles. The prediction module 702 is configured to construct an IDM microscopic following model according to the acceleration, speed and position of the controlled automatic driving vehicle and the acceleration, speed and position of the surrounding traffic vehicles, so as to predict the state of the surrounding traffic vehicles in a preset future long time domain by using the IDM microscopic following model. The planning module 703 is configured to construct an optimization cost function according to the state of the surrounding traffic vehicles in the preset future long time domain, so as to plan the lane-following strategy or the lane-changing strategy of the controlled automatic driving vehicle by using the optimization cost function. The path solving module 704 is configured to solve the reference path of the controlled automatic driving vehicle by using a quintic polynomial based on the lane-following strategy or the lane-changing strategy. The vehicle speed solving module 705 is configured to establish a cost function of the optimal control problem of the continuous intersection predictive cruise control according to the reference path, so as to solve the optimal cruise speed of the driving lane. The control module 706 is configured to send the reference path and the optimal cruise speed of the driving lane to the controlled automatic driving vehicle, so as to control the controlled automatic driving vehicle.

[0128] It should be noted that the foregoing explanation of the embodiment of the predictive cruise and lane-changing decision integrated control method is also applicable to the embodiment of the predictive cruise and lane-changing decision integrated control device, which will not be described here again.

[0129] The predictive cruise and lane-changing decision integrated control device according to the embodiment of the application has the following beneficial effects:

[0130] (1) Combined with the advantages of cloud platform and vehicle platform, based on the wide range of sensing ability and powerful computing advantage of cloud control system, combined with rich traffic prediction information to obtain higher lane change strategy and optimal cruise speed planning, the vehicle platform can obtain higher real-time and more stable controller algorithm with the help of the decision planning of the cloud platform, and execute the vehicle acceleration / deceleration and front wheel steering angle instructions to comprehensively improve the driving economy and safety;

[0131] (2) A predictive cruise and lane change decision based on surrounding vehicle traffic prediction information is designed. First, the IDM model is used to predict the traffic information of the surrounding vehicles in a long time domain, and the artificial potential field method is used to obtain the corresponding following / lane changing strategy, which fully utilizes the super-vision advantage of the cloud platform and effectively improves the application scene of the traditional predictive cruise speed planning. Then, the optimal cruise speed optimizer will obtain the economic driving speed of the lane according to the comprehensive cost optimization function. Compared with the traditional predictive cruise speed planning which is limited to single lane driving and does not consider the predictive cruise control strategy, the controlled vehicle's traffic efficiency is improved, and the economic driving of the controlled vehicle after lane changing is also considered.

[0132] (3) The vehicle platform receives the planning speed and following / lane changing strategy transmitted by the cloud, and designs a vehicle chassis longitudinal and lateral integrated model predictive controller considering obstacle avoidance risk, controlled vehicle state quantity and input quantity constraint. In order to improve the path tracking, speed tracking and obstacle avoidance performance of the designed controller, the weighting coefficients of the controller are optimized, so as to realize the driving economy, smoothness and safety of the controlled vehicle on the multi-signal control urban road.

[0133] FIG. 8 is a structural schematic diagram of an automatic driving vehicle provided by an embodiment of the present application. The automatic driving vehicle can include:

[0134] The memory 801, the processor 802 and the computer program stored in the memory 801 and executable on the processor 802.

[0135] The processor 802 implements the predictive cruise and lane change decision integrated control method provided in the above embodiments when executing the program.

[0136] Further, the automatic driving vehicle further includes:

[0137] The communication interface 803 is used for communication between the memory 801 and the processor 802.

[0138] The memory 801 is used to store the computer program executable on the processor 802.

[0139] The memory 801 can include a high-speed RAM memory, and can also include a non-volatile memory, such as at least one disk memory.

[0140] If the memory 801, the processor 802 and the communication interface 803 are implemented independently, the communication interface 803, the memory 801 and the processor 802 can be connected to each other through a bus and complete communication between each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For convenience of representation, only one thick line is used in FIG. 8, but it does not mean that there is only one bus or only one type of bus.

[0141] Optionally, in a specific implementation, if the memory 801, the processor 802 and the communication interface 803 are integrated on a chip, the memory 801, the processor 802 and the communication interface 803 can complete communication between each other through an internal interface.

[0142] The processor 802 can be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0143] The embodiments of the present application also provide a computer program product, which, when executed by a processor, implements the above-mentioned control method for integrating predictive cruise and lane change decision.

[0144] The embodiments of the present application also provide a computer readable storage medium, which stores a computer program, and the program, when executed by a processor, implements the above-mentioned control method for integrating predictive cruise and lane change decision.

[0145] In the description of the application, reference to "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" means that a particular feature, structure, material, or characteristic being described is included in at least one embodiment or example of the application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples. Moreover, the terminology used in the description is for the purpose of describing the particular versions only and is not intended to be limiting. The use of particular terms to describe the particular versions should not be used to limit the scope of the application to the particular innovative features presented because the novel concepts and embodiments presented herein are capable of other implementations.

[0146] In addition, the terms "first", "second", etc. are used herein only to describe various steps in a method, process, or algorithm. Thus, the terms are used herein to distinguish between two or more steps in the method, process or algorithm and are not necessarily used to describe a sequence or order of steps. In addition, the terms "first", "second", etc. are used herein solely to distinguish between two or more related elements or steps in a process, method or algorithm. Thus, these terms are used herein solely to distinguish between two or more related elements or steps in a process, method or algorithm and are not necessarily used to describe a sequence or order of elements or steps. Further, the use of these terms in the description is not intended to exclude the related elements or steps from the protection automatically afforded by a claim limiting the claim to the elements or steps.

[0147] Any process or method described in a flowchart or otherwise described herein can be understood as representing code modules, segments, or portions of code that include one or more executable instructions for implementing specific logic functions (or steps) of the process, and the various embodiments of the application include additional implementations in which the order of steps can differ from those shown or discussed, including a step can occur at

[0148] The logic and / or steps represented in the flowcharts and / or described herein, for example, can be considered as a sequence of instructions to implement logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, apparatus, or device, such as a computer-based system, processor- based system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For purposes of this specification, a "computer-readable medium" can be any apparatus that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. The computer-readable medium can be a computer- readable storage medium or a computer-readable signal medium. The computer- readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium include the following: an electrical connection having one or more wires (electrical connections), a portable computer diskette (a magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium can even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via, for example, optical scanning of the paper or other medium, then compiled, interpreted, or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.

[0149] It should be understood that aspects of the application can be implemented in hardware, software, firmware or combinations thereof. In the above embodiments, the N steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. If implemented in hardware and in another embodiment, any of the following technologies, known in the art, or their combinations can be used: discrete logic circuitry having logic gates for implementing logic functions on data signals, application specific integrated circuits having appropriate combinational logic gates, programmable gate arrays (PGA), field programmable gate arrays (FPGA), etc.

[0150] Those skilled in the art can understand that all or part of the steps carried out by the above-mentioned embodiment methods can be completed by programs instructing related hardware, and the programs can be stored in a computer-readable storage medium. When the programs are executed, one or a combination of the steps of the method embodiments is included.

[0151] In addition, each of the functional units in the various embodiments of the present application can be integrated in one processing module, or each of the units can be physically present separately, or two or more units can be integrated in one module. The integrated module can be implemented in the form of hardware or in the form of a software functional module. When the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer readable storage medium.

[0152] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it should be understood that the above embodiments are exemplary and should not be construed as limiting the present application, and those skilled in the art can make changes, modifications, replacements and variations to the above embodiments within the scope of the present application.

Claims

1. A control method of predictive cruise and lane change decision integration, characterized in that, The method is applied to a cloud control platform and comprises the following steps: receiving acceleration, speed and position of a controlled automatic driving vehicle and acceleration, speed and position of surrounding traffic vehicles at a current time; constructing an IDM microscopic following model according to the acceleration, speed and position of the controlled automatic driving vehicle and the acceleration, speed and position of the surrounding traffic vehicles to predict a state of the surrounding traffic vehicles in a preset future long time domain by using the IDM microscopic following model; constructing an optimization cost function according to the state of the surrounding traffic vehicles in the preset future long time domain to plan a following strategy or a lane-changing strategy of the controlled automatic driving vehicle by using the optimization cost function; solving a reference path of the controlled automatic driving vehicle by using a quintic polynomial based on the following strategy or the lane-changing strategy; establishing an optimal control problem cost function of a continuous intersection predictive cruise control according to the reference path to solve an optimal cruise speed of a driving lane; sending the reference path and the optimal cruise speed of the driving lane to the controlled automatic driving vehicle to control the controlled automatic driving vehicle.

2. The control method of predictive cruise and lane change decision integration according to claim 1, characterized in that, The expression of the surrounding traffic vehicle state of the preset future long time domain is: where a id (k) is the expected acceleration of the surrounding traffic vehicle, V e is the expected vehicle speed of the surrounding traffic vehicle, a max is the maximum acceleration of the surrounding traffic vehicle, V id (k) is the speed of any surrounding traffic vehicle at time k, Δv id (k) is the longitudinal relative speed of any vehicle to its preceding vehicle at time k, δ a is the acceleration exponent, Δx id (k) is the actual following distance of any vehicle at time k, s * (V id (k), Δv id (k)) is the expected following distance of any vehicle at time k, s0 is the minimum vehicle spacing at rest, THW is the expected vehicle headway, and b is the comfortable deceleration set by the vehicle.

3. The control method of predictive cruise and lane change decision integration according to claim 1, characterized in that, The step of constructing the optimization cost function according to the state of the surrounding traffic vehicles in the future long time domain to plan the following strategy or the lane-changing strategy comprises: generating an optimization cost function combined by a controlled automatic driving vehicle constraint function, an obstacle avoidance risk evaluation function and a function for preventing the controlled vehicle from frequently changing lanes based on the state of the surrounding traffic vehicles in the future long time domain; solving the optimization cost function and planning the following strategy or the lane-changing strategy of the controlled automatic driving vehicle according to a solution of the optimization cost function.

4. The control method of predictive cruise and lane change decision integration according to claim 3, characterized in that, The expression of the optimization cost function is: In the formula, J is the optimization cost function, m is a total mass of the controlled automatic driving vehicle, ω1, ω2 and ω3 are weighting coefficients of each optimization function, r1 is the controlled automatic driving vehicle constraint function, r2 is the obstacle avoidance risk evaluation function, and r3 is the function for preventing the controlled vehicle from frequently changing lanes.

5. The control method of predictive cruise and lane change decision integration according to claim 1, characterized in that, The step of solving the reference path by using the quintic polynomial based on the following strategy or the lane-changing strategy comprises: in a case where the following strategy is adopted, selecting a road center line closest to a center coordinate of the controlled automatic driving vehicle as a reference path for straight driving of a current lane and sending the reference path to the controlled automatic driving vehicle through the cloud control platform; in a case where the lane-changing strategy is adopted, generating a plurality of generalized lane-changing trajectory clusters by using the quintic polynomial, solving coefficients of the quintic polynomial according to a preset constraint condition of longitudinal and lateral motion of the controlled automatic driving vehicle to obtain an optimal lane-changing trajectory as the reference path, and sending the reference path to the controlled automatic driving vehicle.

6. The control method of predictive cruise and lane change decision integration of claim 1, wherein, The step of establishing the optimal control problem cost function of the continuous intersection predictive cruise control according to the optimal following or lane-changing trajectory to solve the optimal cruise speed of the driving lane comprises: based on the optimal following or lane-changing trajectory, establishing a comprehensive energy consumption function, a driving time cost function, a speed fluctuation cost function and an overspeed penalty cost function; According to the comprehensive energy consumption function, the driving time cost function, the speed fluctuation cost function and the overspeed penalty cost function, an optimal control problem cost function of a continuous intersection predictive cruise control is established to solve a driving lane optimal cruise speed, and the driving lane optimal cruise speed is sent to the controlled automatic driving vehicle.

7. The control method of predictive cruise and lane change decision integration according to claim 1, characterized in that, Further comprising: A three-degree-of-freedom vehicle dynamics model of the controlled automatic driving vehicle in longitudinal, lateral and yaw planes is constructed; A dynamics-constrained MPC controller is arranged in the controlled automatic driving vehicle; The reference path and the driving lane optimal cruise speed are taken as reference quantities of the dynamics-constrained MPC controller, so that the controlled automatic driving vehicle is dynamically constrained according to the three-degree-of-freedom vehicle dynamics model in longitudinal, lateral and yaw planes and the reference quantities.

8. The control method of predictive cruise and lane change decision integration according to claim 7, characterized in that, The expression of the three-degree-of-freedom vehicle dynamics model in the longitudinal, lateral, and yaw planes is: wherein m is the total mass of the controlled autonomous vehicle, for longitudinal acceleration, For lateral acceleration, V x For longitudinal Speed, V y is the lateral velocity, and γ is the yaw angular velocity, I z Iz x , F y and M z are the longitudinal, lateral and total yawing force of the controlled autonomous vehicle, respectively.

9. The control method of predictive cruise and lane change decision integration according to claim 7, characterized in that, The reference path and the driving lane optimal cruise speed are taken as reference quantities of the MPC controller, so that the controlled automatic driving vehicle is dynamically constrained according to the three-degree-of-freedom vehicle dynamics model in longitudinal, lateral and yaw planes and the reference quantities, including: When the controlled automatic driving vehicle executes the following strategy, the reference path and the driving lane optimal cruise speed are taken as reference quantities of the MPC controller, an MPC objective function is constructed according to a longitudinal vehicle dynamics model and the reference quantities, so that the longitudinal of the controlled automatic driving vehicle is dynamically constrained; When the controlled automatic driving vehicle executes the lane-changing strategy, the reference path and the driving lane optimal cruise speed are taken as reference quantities of the MPC controller, an MPC objective function is constructed according to a lateral and yaw plane dynamics model and the reference quantities, so that the lateral and yaw plane of the controlled automatic driving vehicle is dynamically constrained.

10. A control device of predictive cruise and lane change decision integration, characterized by, Applied to a cloud control platform, the device comprises: A receiving module is configured to receive acceleration, speed and position of the controlled automatic driving vehicle and acceleration, speed and position of surrounding traffic vehicles at a current time; A prediction module is configured to construct an IDM microscopic following model according to the acceleration, speed and position of the controlled automatic driving vehicle and the acceleration, speed and position of the surrounding traffic vehicles, so as to predict a state of the surrounding traffic vehicles in a preset future long time domain by using the IDM microscopic following model; A planning module is configured to construct an optimization cost function according to the state of the surrounding traffic vehicles in the preset future long time domain, so as to plan a following strategy or a lane-changing strategy of the controlled automatic driving vehicle by using the optimization cost function; A solving path module is configured to solve a reference path of the controlled automatic driving vehicle by using a quintic polynomial based on the following strategy or the lane-changing strategy; A solving speed module is configured to establish an optimal control problem cost function of a continuous intersection predictive cruise control according to the reference path, so as to solve a driving lane optimal cruise speed; A control module is configured to send the reference path and the driving lane optimal cruise speed to the controlled automatic driving vehicle, so as to control the controlled automatic driving vehicle.

11. An autonomous vehicle, comprising: Further comprising: - a memory, a processor, and a computer program stored on the memory and runable on the processor, the processor executing the program to implement the control method of predictive cruise and lane change decision integration according to any one of claims 1-9.

12. A computer program product, characterised in that, The computer program / instructions, when executed by the processor, implement the control method of predictive cruise and lane change decision integration according to any one of claims 1-9.

13. A computer readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor for implementing the control method of predictive cruise and lane change decision integration according to any one of claims 1-9.

Images