CN115257724A

CN115257724A - Safety and energy-saving decision control method and system for plug-in hybrid electric vehicle

Info

Publication number: CN115257724A
Application number: CN202210802285.5A
Authority: CN
Inventors: 赵治国; 李涛
Original assignee: Tongji University
Current assignee: Tongji University
Priority date: 2022-07-07
Filing date: 2022-07-07
Publication date: 2022-11-01

Abstract

The invention relates to a safety and energy-saving decision control method and a system for a plug-in hybrid electric vehicle, wherein the method comprises the following steps: acquiring environmental information around a target vehicle, filtering the obstacle and determining a dangerous obstacle; respectively carrying out track prediction and collision detection on the self-vehicle and the dangerous barrier, and adopting a game decision-making theory to carry out self-vehicle safety behavior decision; based on the decision result of the own vehicle safety behavior, a vehicle longitudinal dynamic state model in a discrete state is established, a multi-objective coordination cost optimization function is constructed according to the vehicle fuel economy, safety and driving comfort with minimum system energy consumption, and a local driving track and a longitudinal safe and economical vehicle speed are planned; and according to the longitudinal safe economic vehicle speed obtained by planning, performing longitudinal vehicle speed following by adopting a model predictive control algorithm to obtain the target required driving torque of the vehicle, and controlling the target vehicle. Compared with the prior art, the plug-in hybrid power energy-saving potential is fully developed on the premise of ensuring the safe running of the vehicle.

Description

Safety and energy-saving decision control method and system for plug-in hybrid electric vehicle

Technical Field

The invention relates to the technical field of energy-saving control of hybrid vehicles, in particular to a safety and energy-saving decision control method and system for a plug-in hybrid vehicle.

Background

The Plug-in Hybrid Electric Vehicle (PHEV) is a new energy Vehicle with the advantages of pure Electric vehicles and Hybrid Electric vehicles, can charge a power battery by utilizing external power grids and engine surplus power, reasonably distributes output energy of Electric energy and mechanical energy in a power system in the driving process, can fully exert the advantages of the Plug-in Hybrid Electric Vehicle configuration, and improves the fuel economy of the Vehicle. The working mode of the power system is closely related to the instantaneous running working condition, and the problems of different modes and switching coordination control under different working conditions need to be considered.

At present, the intelligent safety decision under the instantaneous traffic environment has the problems of low control dimensionality and poor energy-saving effect of energy coordination optimization control, and a novel safety energy-saving decision control method and a novel safety energy-saving decision control system for a plug-in hybrid electric vehicle are urgently needed to be designed so as to fully exploit the energy-saving potential of the plug-in hybrid electric vehicle.

Disclosure of Invention

The invention aims to overcome the defects in the prior art and provides a safety energy-saving decision control method and a safety energy-saving decision control system for a plug-in hybrid electric vehicle, which can fully exploit the potential of energy saving of the plug-in hybrid electric vehicle.

The purpose of the invention can be realized by the following technical scheme:

according to a first aspect of the invention, a safety and energy-saving decision control method for a plug-in hybrid electric vehicle is provided, which comprises the following steps:

s1, obtaining environmental information around a target vehicle, filtering obstacles, and determining dangerous obstacles; respectively carrying out track prediction and collision detection on the self-vehicle and the dangerous barrier, and adopting a game decision-making theory to carry out self-vehicle safety behavior decision;

s2, establishing a vehicle longitudinal dynamic state model in a discrete state based on a self-vehicle safety behavior decision result, constructing a multi-objective coordination cost optimization function according to the vehicle fuel economy, safety and driving comfort with minimum system energy consumption, and planning a local driving track and a longitudinal safe and economical vehicle speed;

and S3, according to the planned longitudinal safe economic vehicle speed, performing longitudinal vehicle speed following by adopting a model predictive control algorithm to obtain a vehicle target required driving torque, and controlling the target vehicle.

Preferably, the step S1 obtains environmental information around the target vehicle, filters the obstacle, and determines the dangerous obstacle, specifically:

1) And (3) performing preliminary filtering according to the existence or nonexistence of the transverse speed of the obstacle relative to the track of the self-vehicle, wherein the preliminary filtering comprises the following steps:

filtering the obstacles with transverse and longitudinal speeds less than a set value and not on the track of the vehicle;

predicting the track based on the position and the speed of the own vehicle and the barrier at the current moment, and filtering when the motion trends of the two vehicles are separated;

2) And determining a filtering area of the self vehicle according to the position, the speed boundary and the movement trend of the obstacle relative to the self vehicle, and performing secondary filtering on the obstacle which is not in the filtering area of the self vehicle and has no transverse speed on the track to determine the dangerous obstacle.

Preferably, the trajectory prediction in step S1 specifically includes:

predicting the track of the vehicle: predicting a running track and a self-vehicle attitude of the self-vehicle in a future period of time according to the current position, the course angle, the driving operation characteristic and the vehicle dynamics characteristic of the target vehicle, wherein the running track and the self-vehicle attitude comprise track points, time for reaching the track points and vehicle attitude;

predicting the track of the obstacle: and judging the behavior and driving style probability of the barrier according to the historical data of the barrier, and predicting the long-term track of the barrier by adopting a multi-source information fusion method to obtain the profile and the transverse and longitudinal speeds of the barrier in a period of time in the future.

Preferably, the collision detection in step S1 specifically includes:

traversing the track predicted positions of all dangerous obstacles and the track predicted position of the self-vehicle, judging whether collision conflict exists from two dimensions of time and distance, and recording obstacle conflict information including the transverse and longitudinal speeds of the obstacles and the time of reaching the conflict point if the collision occurs, and the collision information of the self-vehicle including the time of reaching the conflict point and the distance of the conflict point of the self-vehicle.

Preferably, the step S1 adopts a game decision theory to make the decision on the safety behavior of the vehicle, and specifically includes:

1) When the self-vehicle moves straight, the dangerous barrier moves in front of or at the left side and the right side, and reaches the prediction conflict point in advance of the self-vehicle for a certain time, the self-vehicle does not move in an accelerated way;

2) When the self-vehicle moves straight, the dangerous barrier moves in front of or at the left side and the right side, and reaches the prediction conflict point with the self-vehicle at the same time within a certain time range, and the collision risk exists, the self-vehicle decelerates to drive;

3) When the self-vehicle runs straight, the dangerous barrier runs in front of or at the left side and the right side, no conflict exists with the self-vehicle, and the self-vehicle runs normally;

4) When the self-vehicle turns, the dangerous barrier runs in front of the self-vehicle and reaches the prediction conflict point with the self-vehicle within a certain time range, the collision risk exists, and the self-vehicle decelerates to run.

Preferably, the step S2 specifically includes:

virtualizing dangerous barrier track prediction, track curvature limitation and safety behavior decision results into control targets, constructing a discrete vehicle longitudinal dynamic model, constructing a multi-target coordination cost optimization function by taking vehicle fuel economy, safety and driving comfort as optimization targets, carrying out longitudinal vehicle speed planning point by combining a vehicle running track and an optimal control algorithm under the system constraint condition, updating the target object state in real time, and dynamically updating local longitudinal safe speed planning according to the actual vehicle speed;

when the self-vehicle is subjected to decision of safety behaviors of deceleration and non-acceleration, the local longitudinal safe economic vehicle speed is planned, and the required driving power is intervened in advance.

Preferably, in the vehicle longitudinal dynamic model in step S2, the mathematical expression is:

y(k)＝Cx(k)

wherein x is a state variable, k represents the kth sampling time,

the representative system matrix, y is the system output, C is the output matrix, u is the control input, the speed or acceleration of the vehicle is the control quantity, w is the system disturbance, and the speed or acceleration of the obstacle is the disturbance quantity.

Preferably, the multi-objective coordination cost optimization function in step S2 has an expression:

J＝J_F+J_s+J_c

J_s＝w_ΔdΔd²+w_ΔvΔv²

J_C＝w_aa_f ²

in the formula, J_FOptimizing the performance index, w, for fuel economy_uTo the desired acceleration weight coefficient, w_duA weight coefficient that is a desired rate of acceleration change; j. the design is a square_sFor system security tracking performance index, w_ΔdIs the vehicle distance error weight coefficient, Δ d is the vehicle distance, w_ΔvIs a vehicle relative velocity weight coefficient, Δ v is the relative velocity; j is a unit of_CAs an index of vehicle comfort performance, w_aIs a longitudinal acceleration weight coefficient, a_fIs the longitudinal acceleration.

And the multi-target coordination cost function meets the constraints of vehicle economy, safety and comfort performance as the constraint conditions of comprehensive performance.

Preferably, the step S3 adopts a model predictive control algorithm to perform longitudinal vehicle speed following control, specifically:

selecting a window T in the prediction time domain_p＝[k+1:k+p]Internal output y_pThe square accumulation of the difference value between the (i | k) and the target vehicle speed v (i) is optimized to obtain a control variable sequence with the minimum target function in the corresponding prediction time domain, the first value in the sequence is used as the total required torque value of the current vehicle, and the process is repeated at the next moment;

u_min≤u≤u_max

wherein the target vehicle speed v (i) is a vehicle longitudinal safe economic vehicle speed reference curve, y_p(i | k) output quantities, u, from real-time iterations of the vehicle dynamics model_min、u_maxRepresenting the upper and lower limits of the current demanded torque.

According to a second aspect of the present invention, there is provided a system based on the safety and energy saving decision control method for a plug-in hybrid electric vehicle, the system comprising:

the environment sensing module is used for acquiring environment information around the target vehicle through an intelligent sensor, wherein the environment information comprises obstacle information;

the obstacle filtering module is used for filtering obstacles around the target vehicle to obtain dangerous obstacles;

the track prediction module is used for predicting the track of the dangerous barrier and the vehicle obtained by the barrier filtering module;

the collision detection module is used for carrying out collision detection on the dangerous obstacles obtained by the obstacle filtering module and the self-vehicle;

the safety behavior decision module is used for making a self-vehicle safety behavior decision based on the track prediction and the collision detection result;

the local longitudinal speed planning module is used for planning the speed of the self-vehicle based on the decision result of the safety behavior of the self-vehicle to obtain the longitudinal safe economic speed;

and the speed following control module is used for carrying out longitudinal speed following control on the bicycle.

Compared with the prior art, the invention has the following advantages:

1) The real-time decision control method provided by the invention takes the optimal vehicle economy as a constraint condition in action decision, speed planning and vehicle speed following links on the premise of ensuring the vehicle safety in a dynamic traffic environment, can further improve the fuel economy of the plug-in hybrid electric vehicle, and solves the problems of low intelligent safety decision control dimensionality and poor energy-saving effect of energy coordination optimization control in the current instantaneous traffic environment;

2) The number of collision detection times is reduced by filtering obstacles around the target vehicle for multiple times;

3) When the self-vehicle is subjected to deceleration and non-acceleration safety decisions, the required driving power is intervened in advance by further planning the local longitudinal safe economic vehicle speed, so that energy loss caused by rapid acceleration and rapid deceleration is avoided, the torque of an engine and a motor is reasonably distributed, and the economy of the vehicle is improved;

4) The intelligent sensing technology based on the intelligent sensor provides a way for obtaining short-distance working condition information, and the intelligent auxiliary controller improves the calculation power and provides calculation power support for multi-source information fusion and barrier motion prediction.

Drawings

FIG. 1 is a flow chart of a method of the present invention;

FIG. 2 is a schematic control flow diagram of the present invention;

FIG. 3 is a schematic diagram of trajectory prediction;

FIG. 4 is a schematic view of filtering obstacles according to a movement trend;

fig. 5 is a schematic view of a filtering area of the bicycle.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, shall fall within the scope of protection of the present invention.

Examples

The embodiment of the method is firstly given, and the safety and energy-saving decision control method of the plug-in hybrid electric vehicle comprises the following steps:

s1, acquiring environmental information around a target vehicle, filtering obstacles and determining dangerous obstacles; respectively carrying out track prediction and collision detection on the self-vehicle and the dangerous barrier, and adopting a game decision-making theory to carry out self-vehicle safety behavior decision;

a, filtering obstacles:

1) The preliminary filtering is carried out according to the existence or nonexistence of the transverse speed of the obstacle relative to the track of the self-vehicle, and the preliminary filtering comprises the following steps:

filtering obstacles with transverse and longitudinal speeds less than a set value and not on the track of the vehicle;

predicting the track based on the positions and the speeds of the own vehicle and the barrier at the current moment, and filtering when the motion trends of the two vehicles are separated from each other, as shown in FIG. 4;

2) The current position of the vehicle is taken as a reference, the size of the vehicle filtering area is determined according to the transverse offset of the vehicle track, the vehicle course angle, the intersection mark, the lane mark and the like, the vehicle filtering area is determined, and as shown in fig. 5, the obstacle which is outside the vehicle filtering area and has no transverse speed on the track is filtered again, so that the collision detection times are reduced.

B, track prediction:

predicting the track of the bicycle: predicting a running track and a self-vehicle attitude of the self-vehicle in a future period of time according to the current position, the course angle, the driving operation characteristic and the vehicle dynamics characteristic of the target vehicle, wherein the running track and the self-vehicle attitude comprise track points, time for reaching the track points and vehicle attitude;

obstacle trajectory prediction: and judging the behavior and driving style probability of the barrier according to the historical data of the barrier, and performing long-term prediction on the track of the barrier by adopting a multi-source information fusion method to obtain the profile and the transverse and longitudinal speeds of the barrier in a period of time in the future.

C, collision detection:

traversing the track predicted positions of all dangerous obstacles and the track predicted positions of the self-vehicle, judging whether collision conflict exists from two dimensions of time and distance, and recording obstacle conflict information including the transverse and longitudinal speeds of the obstacles and the time of reaching the conflict point if the collision occurs, and the collision information of the self-vehicle including the time of the self-vehicle reaching the conflict point and the distance of the conflict point.

D, decision of safety behavior of the own vehicle, as shown in table 1:

TABLE 1

S2, virtualizing dangerous barrier track prediction, track curvature limitation and safety behavior decision results into control targets based on own vehicle safety behavior decision results, constructing a discrete vehicle longitudinal dynamic model, constructing a multi-target coordination cost optimization function by taking vehicle fuel economy, safety and driving comfort as optimization targets, carrying out longitudinal vehicle speed planning point by combining own vehicle running tracks and an optimal control algorithm under the system constraint condition, updating the target object state in real time, and dynamically updating local longitudinal safe speed planning according to actual vehicle speed;

The vehicle longitudinal dynamics model has the mathematical expression as follows:

y(k)＝Cx(k)

wherein x is a state variable, k represents the kth sampling time,

the system matrix is represented, y is the system output, C is the output matrix, u is the control input, the speed or acceleration of the vehicle is used as the control quantity, w is the system disturbance, and the speed or acceleration of the obstacle is used as the disturbance quantity.

The multi-objective coordination cost optimization function has the expression:

J＝J_F+J_s+J_c

J_s＝w_ΔdΔd²+w_ΔvΔv²

J_C＝w_aa_f ²

in the formula, J_FOptimizing the performance index, w, for fuel economy_uTo the desired acceleration weight coefficient, w_duA weight coefficient that is a desired rate of acceleration change; j. the design is a square_sFor system security tracking performance index, w_ΔdIs the vehicle distance error weight coefficient, Δ d is the vehicle distance, w_ΔvIs a vehicle relative speed weight coefficient, and deltav is a relative speed; j. the design is a square_CAs an index of vehicle comfort performance, w_aIs a longitudinal acceleration weight coefficient, a_fIs the longitudinal acceleration.

S3, according to the planned longitudinal safe economic vehicle speed, performing longitudinal vehicle speed following by adopting a model predictive control algorithm to obtain a vehicle target required driving torque, and controlling a target vehicle, wherein the method specifically comprises the following steps:

the longitudinal vehicle speed following control is carried out by adopting a model predictive control algorithm, which specifically comprises the following steps:

selecting a window T in the prediction time domain_p＝[k+1:k+p]Internal output y_pThe square accumulation of the difference value between the (i | k) and the target vehicle speed v (i) is optimized to obtain the control with the minimum target function in the corresponding prediction time domainMaking a variable sequence, taking a first value in the sequence as a total required torque value of the current self vehicle, and repeating the process at the next moment;

u_min≤u≤u_max

The method of the present embodiment will be described in detail below with reference to the drawings.

As shown in fig. 2:

(1) Predicting the track of the bicycle:

and predicting the running track of the vehicle according to the operation of a driver (accelerator opening, brake opening, steering lamps and steering wheel turning angles) and the feedback of the vehicle state (navigation line, vehicle speed and lane line). And on the basis of the vehicle dynamics, keeping the current vehicle speed unchanged, and predicting the vehicle posture on a future track, wherein the vehicle posture comprises the information of the profile, the arrival time and the like of the vehicle at each track point.

(2) Obstacle trajectory prediction:

firstly, filtering obstacles, acquiring local road condition information such as distance, speed and position of obstacles around a vehicle based on an intelligent sensor, and filtering the obstacles, wherein V _Selfrepresents the speed of the vehicle and V _ OBS is an obstacle vector speed, as shown in FIG. 3. Firstly, filtering obstacles OBS _2, OBS _4 and OBS _5 which are outside an area and have no transverse speed on a track according to a static obstacle filtering mode; and then predicting the running track according to the self-vehicle state in a dynamic barrier filtering mode, predicting the track according to the position and the speed of the barrier at the current moment, and filtering the barrier OBS _3 when the two movement trends are separated.

Then, the attitude of the obstacle that does not meet the obstacle filtering rule is predicted, as shown in fig. 3, the obstacle OBS _1, and taking the state of the obstacle at time t0 as an example, the initial position of the obstacle is (x 0, y 0), the current speed of the obstacle is (vx, vy), and after time t, the new obstacle position (x 1, y 1) can be expressed as the following formula, so that the coordinates and attitude of each position of the obstacle in the prediction time can be calculated.

(3) Collision detection:

according to the barrier filtering result, traversing all dangerous barrier track prediction positions and the self-vehicle track prediction position, judging whether collision conflict exists from two dimensions of time and distance, and recording information such as the transverse and longitudinal speed of the barrier, the time of reaching the conflict point by the self-vehicle, the distance of the conflict point and the like if the collision occurs.

(4) And (4) safety decision making:

and adopting a road right and game decision theory, and adopting the following safety decision behavior for collision information of the vehicle and the barrier. When the self-vehicle moves straight, the most dangerous barrier moves in front of or at the left side and the right side, and reaches the prediction conflict point in advance of the self-vehicle by a certain time, the self-vehicle does not move in an accelerated way.

(5) Local longitudinal velocity planning:

the planned vehicle speed trajectory of the host vehicle is shown in FIG. 3, and the vehicle speed at the t-th time is P (x)_t,y_t,v_t) The required driving power is intervened in advance by planning the local economic vehicle speed, energy loss caused by rapid acceleration and rapid deceleration is avoided, the torque of the engine and the motor is distributed reasonably, and the economical efficiency of the vehicle is improved.

(6) Track following:

according to the planned local economic speed, in order to enable the vehicle to track the expected track smoothly, safely and accurately, a model prediction control algorithm is adopted for follow-up control. Selecting in a prediction temporal window (T)_p＝[k+1:k+p]) Internal output y_pThe square accumulation of the difference value between the (i | k) and the target track r (i) is carried out, and the target function in the corresponding prediction time domain is obtained through an optimization algorithmAnd counting the minimum control variable sequence, taking the first value as the total required torque value of the whole vehicle at present, and repeating the process at the next moment.

u_min≤u≤u_max

In the formula, the target speed v (i) is a reference vehicle speed curve of a vehicle speed track plan, y_p(i | k) is obtained by real-time iteration of the vehicle dynamics model, u_min、u_maxRepresenting the upper and lower limits of the current demanded torque.

Next, a system embodiment of the present invention is given, in which a system based on the plug-in hybrid electric vehicle safety and energy saving decision control method includes:

the environment sensing module is used for acquiring environment information around the target vehicle through an intelligent sensor, wherein the environment information comprises barrier information;

the local longitudinal speed planning module is used for planning the speed of the self-vehicle based on the decision result of the safety behavior of the self-vehicle to obtain the longitudinal safe economic vehicle speed;

and the speed following control module is used for carrying out following control on the longitudinal speed of the self vehicle.

While the invention has been described with reference to specific embodiments, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A safety and energy-saving decision control method for a plug-in hybrid electric vehicle is characterized by comprising the following steps:

s2, establishing a vehicle longitudinal dynamic state model in a discrete state based on a self vehicle safety behavior decision result, constructing a multi-target coordination cost optimization function according to the vehicle fuel economy, safety and driving comfort with minimum system energy consumption, and planning a local driving track and a longitudinal safe economic vehicle speed;

2. The safety and energy saving decision control method for the plug-in hybrid electric vehicle according to claim 1, wherein the step S1 is performed by acquiring environmental information around the target vehicle, filtering obstacles, and determining dangerous obstacles, specifically:

3. The safety and energy saving decision control method for the plug-in hybrid electric vehicle according to claim 1, wherein the trajectory prediction in the step S1 specifically comprises:

4. The safety and energy saving decision control method for the plug-in hybrid electric vehicle according to claim 1, wherein the collision detection in the step S1 specifically comprises:

5. The safety and energy conservation decision control method for the plug-in hybrid electric vehicle according to claim 1, wherein a game decision theory is adopted in the step S1 to make a decision on the safety behavior of the vehicle, and specifically comprises the following steps:

6. The safety and energy saving decision control method for the plug-in hybrid electric vehicle according to claim 1, wherein the step S2 specifically comprises:

7. The safety and energy saving decision control method for the plug-in hybrid electric vehicle according to claim 6, wherein the mathematical expression of the vehicle longitudinal dynamics model in the step S2 is as follows:

y(k)＝Cx(k)

wherein x is a state variable, k represents the kth sampling time,

8. The safety and energy saving decision control method for the plug-in hybrid electric vehicle according to claim 6, wherein the multi-objective coordination cost optimization function in the step S2 has an expression as follows:

J＝J_F+J_s+J_c

J_s＝w_ΔdΔd²+w_ΔvΔv²

J_C＝w_aa_f ²

in the formula, J_FOptimizing the performance index, w, for fuel economy_uTo the desired acceleration weight coefficient, w_duA weight factor that is a desired rate of acceleration change; j. the design is a square_sFor system security tracking performance index, w_ΔdIs the vehicle distance error weight coefficient, Δ d is the vehicle distance, w_ΔvIs a vehicle relative speed weight coefficient, and deltav is a relative speed; j is a unit of_CAs an index of vehicle comfort performance, w_aIs a longitudinal acceleration weight coefficient, a_fIs the longitudinal acceleration.

9. The safety and energy-saving decision control method for the plug-in hybrid electric vehicle according to claim 1, wherein a model predictive control algorithm is adopted in the step S3 for longitudinal vehicle speed following control, specifically:

u_min≤u≤u_max

10. A system of a safety and energy saving decision control method for a plug-in hybrid electric vehicle according to claim 1, wherein the system comprises: