A plug-in hybrid vehicle safety and energy-saving decision control method and system
Through game decision theory and model prediction control algorithm, the longitudinal dynamic state of plug-in hybrid vehicles is optimized, which solves the problems of low intelligent safety decision control dimensions and poor energy coordination optimization effects in instantaneous traffic environments, and achieves more efficient fuel fuel Economy and vehicle comfort.
Patent Information
- Application Number
- CN202210802285.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-07-07
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2042-07-07
AI Technical Summary
Existing plug-in hybrid vehicles have low intelligent safety decision-making control dimensions and poor energy coordination and optimization effects in instantaneous traffic environments, and have failed to fully tap the energy saving potential of hybrid power.
Using game decision theory combined with model prediction and control algorithm, multi-objective coordinated cost optimization function is constructed by obtaining environmental information, obstacle filtering, trajectory prediction and collision detection, vertical safety and economic vehicle speed are planned, and vehicle speed follow-up control is carried out to optimize the energy of the power system. distribute.
On the premise of ensuring vehicle safety, the fuel economy of plug-in hybrid vehicles is improved, energy losses caused by sudden acceleration and rapid deceleration are reduced, and the economy and comfort of the vehicle are improved.
Smart Images

Figure CN115257724B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of hybrid vehicle energy-saving control, and in particular to a plug-in hybrid vehicle safety and energy-saving decision-making control method and system. Background Art
[0002] Plug-in Hybrid Electric Vehicle (PHEV) is a new energy vehicle that combines the advantages of pure electric and hybrid vehicles. It can use the external power grid and the surplus power of the engine to charge the power battery. During driving, the output energy of electrical energy and mechanical energy in the power system is reasonably distributed, which can give full play to the advantages of the plug-in hybrid configuration and improve the fuel economy of the vehicle. The working mode of the power system is closely related to the instantaneous driving conditions, and it is necessary to consider the different modes under different conditions and the coordinated control of their switching.
[0003] At present, intelligent safety decision-making in instantaneous traffic environments has the problems of low control dimension and poor energy-saving effect of energy coordination optimization control. It is urgent to design a new plug-in hybrid vehicle safety and energy-saving decision-making control method and system to fully tap the energy-saving potential of plug-in hybrid vehicles. Summary of the invention
[0004] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and to provide a plug-in hybrid vehicle safety and energy-saving decision-making control method and system that can fully tap the energy-saving potential of plug-in hybrid vehicles.
[0005] The purpose of the present invention can be achieved by the following technical solutions:
[0006] According to a first aspect of the present invention, a plug-in hybrid vehicle safety and energy-saving decision control method is provided, the method comprising the following steps:
[0007] Step S1, obtaining environmental information around the target vehicle, filtering obstacles, and determining dangerous obstacles; performing trajectory prediction and collision detection on the ego vehicle and dangerous obstacles respectively, and using game decision theory to make decisions on the ego vehicle's safety behavior;
[0008] Step S2: Based on the decision result of the vehicle safety behavior, a vehicle longitudinal dynamic state model under discrete state is established, a multi-objective coordinated cost optimization function is constructed with the minimum system energy consumption, vehicle fuel economy, safety and driving comfort, and the local driving trajectory and longitudinal safety and economic speed are planned;
[0009] Step S3: Based on the planned longitudinal safe and economical vehicle speed, a model predictive control algorithm is used to perform longitudinal vehicle speed following, obtain the vehicle target demand driving torque, and control the target vehicle.
[0010] Preferably, in step S1, the environmental information around the target vehicle is obtained, obstacles are filtered, and dangerous obstacles are determined, specifically:
[0011] 1) Perform preliminary filtering based on whether the obstacle has a lateral speed relative to the vehicle's trajectory, including:
[0012] Filter obstacles whose lateral and longitudinal speeds are less than the set values and are not on the vehicle's trajectory;
[0013] Based on the current position and speed of the vehicle and the obstacle, the trajectory is predicted. If the two vehicles move in different directions, they are filtered out.
[0014] 2) According to the position, speed boundary and movement trend of the obstacle relative to the ego vehicle, the ego vehicle filtering area is determined. Obstacles outside the ego vehicle filtering area and without lateral speed on the trajectory are filtered again to determine dangerous obstacles.
[0015] Preferably, the trajectory prediction in step S1 is specifically:
[0016] Ego vehicle trajectory prediction: Based on the current position, heading angle, driving operation characteristics and vehicle dynamics characteristics of the target vehicle, the driving trajectory and attitude of the ego vehicle in the future are predicted, including the trajectory point, the time to reach the trajectory point and the vehicle attitude;
[0017] Obstacle trajectory prediction: The obstacle behavior and driving style probability are judged based on the historical obstacle data, and the multi-source information fusion method is used to perform long-term obstacle trajectory prediction to obtain the obstacle contour and lateral and longitudinal speed in the future.
[0018] Preferably, the collision detection in step S1 is specifically:
[0019] Traverse the predicted trajectory positions of all dangerous obstacles and the predicted trajectory positions of the ego vehicle, and determine whether there is a collision conflict from the two dimensions of time and distance. If a conflict occurs, record the obstacle conflict information, including the obstacle's lateral and longitudinal speeds and the time it arrives at the conflict point, as well as the ego vehicle conflict information, including the time the ego vehicle arrives at the conflict point and the distance from the conflict point.
[0020] Preferably, in step S1, game decision theory is used to make a decision on the vehicle's safety behavior, specifically:
[0021] 1) When the vehicle is moving straight ahead and a dangerous obstacle is moving in front or on the left or right side and arrives at the predicted conflict point a certain time earlier than the vehicle, the vehicle does not accelerate;
[0022] 2) When the vehicle is traveling straight ahead and a dangerous obstacle is traveling in front or on the left or right side, and reaches the predicted conflict point at the same time as the vehicle within a certain time range, and there is a risk of collision, the vehicle will slow down;
[0023] 3) When the vehicle is moving straight ahead, dangerous obstacles are moving in front or on the left or right sides, and there is no conflict with the vehicle, so the vehicle moves normally;
[0024] 4) When the ego vehicle turns, a dangerous obstacle is moving ahead and reaches the predicted conflict point at the same time as the ego vehicle within a certain time range, and there is a risk of collision, so the ego vehicle slows down.
[0025] Preferably, the step S2 is specifically:
[0026] The prediction of dangerous obstacle trajectories, trajectory curvature restrictions, and safety behavior decision results are virtualized as control targets, and a discrete vehicle longitudinal dynamics model is constructed. With vehicle fuel economy, safety, and driving comfort as optimization targets, a multi-objective coordinated cost optimization function is constructed. Under system constraints, the vehicle's running trajectory and the optimal control algorithm are combined to perform longitudinal vehicle speed planning point by point, and the target object state is updated in real time. The local longitudinal safety speed planning is dynamically updated according to the actual vehicle speed.
[0027] When the vehicle makes a safety behavior decision of deceleration or non-acceleration, the local longitudinal safe and economical speed is planned to intervene in the driving demand power in advance.
[0028] Preferably, the vehicle longitudinal dynamics model in step S2 is mathematically expressed as:
[0029]
[0030] y(k)=Cx(k)
[0031] In the formula, x is the state variable, k represents the kth sampling time, represents the system matrix, y is the system output, C is the output matrix, u is the control input, with the vehicle speed or acceleration as the control variable, and w is the system disturbance, with the obstacle speed or acceleration as the disturbance variable.
[0032] Preferably, the multi-objective coordination cost optimization function in step S2 is expressed as:
[0033] J=J F +J s +J c
[0034]
[0035] J s =w Δd Δd 2 +w Δv Δv 2
[0036] J C =w a af 2
[0037] In the formula, J F Optimize performance indicators for fuel economy, w u is the expected acceleration weight coefficient, w du is the weight coefficient of the expected acceleration change rate; J s Tracking performance indicators for system security, w Δd is the vehicle spacing error weight coefficient, Δd is the vehicle spacing, w Δv is the vehicle relative speed weight coefficient, Δv is the relative speed; J C is the vehicle comfort performance index, w a is the longitudinal acceleration weight coefficient, a f is the longitudinal acceleration.
[0038] Among them, the multi-objective coordination cost function satisfies the constraints of vehicle economy, safety and comfort performance as constraints of comprehensive performance.
[0039] Preferably, in step S3, a model predictive control algorithm is used to perform longitudinal vehicle speed following control, specifically:
[0040] Select the prediction time window T p = output y in [k+1:k+p] p The square of the difference between (i|k) and the target vehicle speed v(i) is accumulated to optimize the control variable sequence with the minimum objective function in the corresponding prediction time domain, and the first value in the sequence is used as the total required torque value of the current vehicle, and this process is repeated at the next moment;
[0041]
[0042] u min ≤u≤u max
[0043] Where, the target speed v(i) is the longitudinal safety and economic speed reference curve of the vehicle, and y p (i|k) is the output obtained by real-time iteration of the vehicle dynamics model, u min 、u max Indicates the upper and lower limits of the current required torque.
[0044] According to a second aspect of the present invention, a system based on the plug-in hybrid vehicle safety and energy-saving decision control method is provided, the system comprising:
[0045] Environmental perception module, used to obtain environmental information around the target vehicle, including obstacle information, through intelligent sensors;
[0046] The obstacle filtering module is used to filter the obstacles around the target vehicle to obtain dangerous obstacles;
[0047] The trajectory prediction module is used to predict the trajectory of the dangerous obstacles and the ego vehicle obtained by the obstacle filtering module;
[0048] A collision detection module is used to perform collision detection between the dangerous obstacles obtained by the obstacle filtering module and the vehicle;
[0049] Safety behavior decision module, used to make safety behavior decisions for the ego vehicle based on trajectory prediction and collision detection results;
[0050] The local longitudinal speed planning module is used to plan the vehicle speed based on the vehicle safety behavior decision results to obtain the longitudinal safe and economical vehicle speed;
[0051] The speed following control module is used to perform longitudinal speed following control of the vehicle.
[0052] Compared with the prior art, the present invention has the following advantages:
[0053] 1) The real-time decision-making control method proposed in the present invention, under the premise of ensuring vehicle safety in a dynamic traffic environment, takes the optimal vehicle economy as a constraint condition in the behavior decision-making, speed planning and speed following links, which can further improve the fuel economy of plug-in hybrid vehicles and solve the problems of low dimensionality of intelligent safety decision-making control and poor energy-saving effect of energy coordinated optimization control in the current instantaneous traffic environment;
[0054] 2) Reduce the number of collision detections by filtering obstacles around the target vehicle multiple times;
[0055] 3) When the vehicle makes a deceleration or non-acceleration safety decision, the local longitudinal safety and economic speed is further planned, the driving demand power is intervened in advance, the energy loss caused by sudden acceleration and deceleration is avoided, the engine and motor torque are reasonably allocated, and the economy of the vehicle is improved;
[0056] 4) The intelligent sensing technology based on smart sensors provides a way to obtain short-distance working condition information. The improvement of the computing power of the intelligent auxiliary controller provides computing power support for multi-source information fusion and obstacle motion prediction. BRIEF DESCRIPTION OF THE DRAWINGS
[0057] Figure 1 is a flow chart of the method of the present invention;
[0058] Figure 2 It is a control flow diagram of the present invention;
[0059] Figure 3 Schematic diagram of trajectory prediction;
[0060] Figure 4 Schematic diagram for filtering obstacles based on movement trends;
[0061] Figure 5 Schematic diagram of the ego vehicle filtering area. DETAILED DESCRIPTION
[0062] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.
[0063] Example
[0064] First, a method embodiment of the present invention is given, which is a plug-in hybrid vehicle safety and energy-saving decision control method, and the method includes the following steps:
[0065] Step S1, obtaining environmental information around the target vehicle, filtering obstacles, and determining dangerous obstacles; performing trajectory prediction and collision detection on the ego vehicle and dangerous obstacles respectively, and using game decision theory to make decisions on the ego vehicle's safety behavior;
[0066] A Obstacle Filtering:
[0067] 1) Perform preliminary filtering based on whether the obstacle has a lateral speed relative to the vehicle's trajectory, including:
[0068] Filter obstacles whose lateral and longitudinal speeds are less than the set values and are not on the vehicle's trajectory;
[0069] Based on the current position and speed of the vehicle and the obstacle, the trajectory is predicted. When the movement trends of the two vehicles are separated, they are filtered out, such as Figure 4 As shown;
[0070] 2) Taking the current position of the vehicle as a reference, determine the size of the ego vehicle filtering area according to the lateral offset of the ego vehicle trajectory, the vehicle heading angle, the intersection sign, the lane sign, etc., and determine the ego vehicle filtering area, such as Figure 5 As shown, obstacles outside the ego vehicle filtering area and without lateral velocity on the trajectory are filtered again, thereby reducing the number of collision detections.
[0071] B trajectory prediction:
[0072] Ego vehicle trajectory prediction: Based on the current position, heading angle, driving operation characteristics and vehicle dynamics characteristics of the target vehicle, the driving trajectory and attitude of the ego vehicle in the future are predicted, including the trajectory point, the time to reach the trajectory point and the vehicle attitude;
[0073] Obstacle trajectory prediction: The obstacle behavior and driving style probability are judged based on the historical obstacle data, and the multi-source information fusion method is used to perform long-term obstacle trajectory prediction to obtain the obstacle contour and lateral and longitudinal speed in the future.
[0074] C Collision Detection:
[0075] Traverse the predicted trajectory positions of all dangerous obstacles and the predicted trajectory positions of the ego vehicle, and determine whether there is a collision conflict from the two dimensions of time and distance. If a conflict occurs, record the obstacle conflict information, including the obstacle's lateral and longitudinal speeds and the time it arrives at the conflict point, as well as the ego vehicle conflict information, including the time the ego vehicle arrives at the conflict point and the distance from the conflict point.
[0076] D. The safety behavior decision of the vehicle is shown in Table 1:
[0077] Table 1
[0078]
[0079] Step S2: Based on the result of the self-vehicle safety behavior decision, the prediction of the trajectory of dangerous obstacles, the trajectory curvature limit, and the safety behavior decision result are virtualized as control targets, a discrete vehicle longitudinal dynamics model is constructed, and the vehicle fuel economy, safety, and driving comfort are taken as optimization targets. A multi-objective coordinated cost optimization function is constructed. Under the system constraints, the self-vehicle running trajectory and the optimal control algorithm are combined to perform longitudinal vehicle speed planning point by point, and the target object state is updated in real time. The local longitudinal safety speed planning is dynamically updated according to the actual vehicle speed.
[0080] When the vehicle makes a safety behavior decision of deceleration or non-acceleration, the local longitudinal safe and economical speed is planned to intervene in the driving demand power in advance.
[0081] The vehicle longitudinal dynamics model, mathematical expression is:
[0082]
[0083] y(k)=Cx(k)
[0084] In the formula, x is the state variable, k represents the kth sampling time, represents the system matrix, y is the system output, C is the output matrix, u is the control input, with the vehicle speed or acceleration as the control variable, and w is the system disturbance, with the obstacle speed or acceleration as the disturbance variable.
[0085] The multi-objective coordination cost optimization function is expressed as:
[0086] J=J F +J s +J c
[0087]
[0088] J s =w Δd Δd 2 +w Δv Δv 2
[0089] J C =w a a f 2
[0090] In the formula, J F Optimize performance indicators for fuel economy, w u is the expected acceleration weight coefficient, w du is the weight coefficient of the expected acceleration change rate; J s Tracking performance indicators for system security, w Δd is the vehicle spacing error weight coefficient, Δd is the vehicle spacing, w Δv is the vehicle relative speed weight coefficient, Δv is the relative speed; J C is the vehicle comfort performance index, w a is the longitudinal acceleration weight coefficient, a f is the longitudinal acceleration.
[0091] Among them, the multi-objective coordination cost function satisfies the constraints of vehicle economy, safety and comfort performance as constraints of comprehensive performance.
[0092] Step S3: Based on the planned longitudinal safe and economical vehicle speed, a model predictive control algorithm is used to perform longitudinal vehicle speed following, and the vehicle target demand driving torque is obtained, and the target vehicle is controlled, specifically:
[0093] The model predictive control algorithm is used for longitudinal vehicle speed following control, specifically:
[0094] Select the prediction time window T p = output y in [k+1:k+p] p The square of the difference between (i|k) and the target vehicle speed v(i) is accumulated to optimize the control variable sequence with the minimum objective function in the corresponding prediction time domain, and the first value in the sequence is used as the total required torque value of the current vehicle, and this process is repeated at the next moment;
[0095]
[0096] u min ≤u≤u max
[0097] Where, the target speed v(i) is the longitudinal safety and economic speed reference curve of the vehicle, and y p (i|k) is the output obtained by real-time iteration of the vehicle dynamics model, u min 、u max Indicates the upper and lower limits of the current required torque.
[0098] Next, the method of this embodiment is described in detail with reference to the accompanying drawings.
[0099] like Figure 2 As shown:
[0100] (1) Vehicle trajectory prediction:
[0101] The vehicle's trajectory is predicted based on the driver's operation (accelerator opening, brake opening, turn signal, steering wheel angle) and vehicle status feedback (navigation route, vehicle speed, lane line). Based on the vehicle's dynamic characteristics, the current vehicle speed is kept unchanged and the vehicle's posture on the future trajectory is predicted, including the vehicle's profile at each trajectory point, arrival time and other information.
[0102] (2) Obstacle trajectory prediction:
[0103] First, obstacle filtering is performed. Based on the intelligent sensor, local road condition information such as the distance, speed, and position of obstacles around the vehicle is obtained, and obstacles are filtered. Figure 3 , V_Self represents the speed of the ego vehicle, and V_OBS is the obstacle vector speed. First, according to the static obstacle filtering method, obstacles OBS_2, OBS_4, and OBS_5 that are outside the area and have no lateral speed on the trajectory are filtered; then according to the dynamic obstacle filtering method, the running trajectory is predicted according to the state of the ego vehicle, and the trajectory is predicted according to the position and speed of the obstacle at the current moment. When the two motion trends are separated, the obstacle OBS_3 is filtered out.
[0104] Then, the posture of obstacles that do not meet the obstacle filtering rules is predicted, such as Figure 3 For obstacle OBS_1, take the state of the obstacle at time t0 as an example, the initial position of the obstacle is (x0, y0), the current speed of the obstacle is (vx, vy), after time t, the new obstacle position (x1, y1) can be expressed as the following formula, so the coordinates and posture of each position of the obstacle within the prediction time can be calculated.
[0105]
[0106] (3) Collision detection:
[0107] According to the obstacle filtering results, the predicted positions of all dangerous obstacles and the predicted positions of the ego vehicle are traversed to determine whether there is a collision conflict from the two dimensions of time and distance. If a conflict occurs, the obstacle's lateral and longitudinal speeds, the time to reach the conflict point, the time the ego vehicle arrives at the conflict point, the distance to the conflict point, and other information are recorded.
[0108] (4) Safety Decision-making:
[0109] Using the right of way and game decision theory, the following safety decision behavior is adopted for the collision information between the ego vehicle and the obstacle. When the ego vehicle is moving straight, the most dangerous obstacle is moving in front or on the left and right sides, and arrives at the predicted conflict point a certain time earlier than the ego vehicle, then the ego vehicle does not accelerate.
[0110] (5) Local longitudinal speed planning:
[0111] The speed trajectory planned by the vehicle is as follows: Figure 3 As shown, the vehicle speed at time t is P(x t ,y t ,v t ), by planning the local economic speed, intervening in driving demand power in advance, avoiding energy loss caused by sudden acceleration and deceleration, and reasonably allocating engine and motor torque to improve the economy of the vehicle.
[0112] (6) Track following:
[0113] According to the planned local economic speed, in order to enable the vehicle to smoothly, safely and accurately track the desired trajectory, the model predictive control algorithm is used for following control. p = output y within [k+1:k+p]) p The square of the difference between (i|k) and the target trajectory r(i) is accumulated, and the control variable sequence with the minimum objective function in the corresponding prediction time domain is obtained through the optimization algorithm. The first value is used as the total required torque value of the current vehicle, and this process is repeated at the next moment.
[0114]
[0115] u min ≤u≤u max
[0116] Where, the target speed v(i) is the reference speed curve for speed trajectory planning, and y p (i|k) is obtained by real-time iteration of the vehicle dynamics model, u min 、u max Indicates the upper and lower limits of the current required torque.
[0117] Next, a system embodiment of the present invention is given, which is a system based on the plug-in hybrid vehicle safety and energy-saving decision control method, and the system includes:
[0118] Environmental perception module, used to obtain environmental information around the target vehicle, including obstacle information, through intelligent sensors;
[0119] The obstacle filtering module is used to filter the obstacles around the target vehicle to obtain dangerous obstacles;
[0120] The trajectory prediction module is used to predict the trajectory of the dangerous obstacles and the ego vehicle obtained by the obstacle filtering module;
[0121] A collision detection module is used to perform collision detection between the dangerous obstacles obtained by the obstacle filtering module and the vehicle;
[0122] Safety behavior decision module, used to make safety behavior decisions for the ego vehicle based on trajectory prediction and collision detection results;
[0123] The local longitudinal speed planning module is used to plan the vehicle speed based on the vehicle safety behavior decision results to obtain the longitudinal safe and economical vehicle speed;
[0124] The speed following control module is used to perform longitudinal speed following control of the vehicle.
[0125] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.
Claims
1. A plug-in hybrid vehicle safety and energy-saving decision control method, characterized in that: The method comprises the following steps: Step S1, obtaining environmental information around the target vehicle, filtering obstacles, and determining dangerous obstacles; performing trajectory prediction and collision detection on the ego vehicle and dangerous obstacles respectively, and using game decision theory to make decisions on the ego vehicle's safety behavior; Step S2: Based on the decision result of the vehicle safety behavior, a vehicle longitudinal dynamic state model under discrete state is established, a multi-objective coordinated cost optimization function is constructed with the minimum system energy consumption, vehicle fuel economy, safety and driving comfort, and the local driving trajectory and longitudinal safety and economic speed are planned; The multi-objective coordination cost optimization function is expressed as: J=J F +J s +J c J s =w Δd Δd 2 +w Δv Δv 2 J C =w a a f 2 In the formula, J F Optimize performance indicators for fuel economy, w u is the expected acceleration weight coefficient, w du is the weight coefficient of the expected acceleration change rate; J s Tracking performance indicators for system security, w Δd is the vehicle spacing error weight coefficient, Δd is the vehicle spacing, w Δv is the vehicle relative speed weight coefficient, Δv is the relative speed; J C is the vehicle comfort performance index, w a is the longitudinal acceleration weight coefficient, a f is the longitudinal acceleration; Among them, the multi-objective coordination cost function satisfies the constraints of vehicle economy, safety and comfort performance as the constraints of comprehensive performance; Step S3: Based on the planned longitudinal safe and economical vehicle speed, a model predictive control algorithm is used to perform longitudinal vehicle speed following, obtain the vehicle target demand driving torque, and control the target vehicle.
2. A plug-in hybrid vehicle safety and energy-saving decision control method according to claim 1, characterized in that: In step S1, the environmental information around the target vehicle is obtained, obstacles are filtered, and dangerous obstacles are determined, specifically: 1) Perform preliminary filtering based on whether the obstacle has a lateral speed relative to the vehicle's trajectory, including: Filter obstacles whose lateral and longitudinal speeds are less than the set values and are not on the vehicle's trajectory; Based on the current position and speed of the vehicle and the obstacle, the trajectory is predicted. If the two vehicles move in different directions, they are filtered out. 2) According to the position, speed boundary and movement trend of the obstacle relative to the ego vehicle, the ego vehicle filtering area is determined. Obstacles outside the ego vehicle filtering area and without lateral speed on the trajectory are filtered again to determine dangerous obstacles.
3. The plug-in hybrid vehicle safety and energy-saving decision control method according to claim 1, characterized in that: The trajectory prediction in step S1 is specifically as follows: Ego vehicle trajectory prediction: Based on the current position, heading angle, driving operation characteristics and vehicle dynamics characteristics of the target vehicle, the driving trajectory and attitude of the ego vehicle in the future are predicted, including the trajectory point, the time to reach the trajectory point and the vehicle attitude; Obstacle trajectory prediction: The obstacle behavior and driving style probability are judged based on the historical obstacle data, and the multi-source information fusion method is used to perform long-term obstacle trajectory prediction to obtain the obstacle contour and lateral and longitudinal speed in the future.
4. The plug-in hybrid vehicle safety and energy-saving decision control method according to claim 1, characterized in that: The collision detection in step S1 is specifically as follows: Traverse the predicted trajectory positions of all dangerous obstacles and the predicted trajectory positions of the ego vehicle, and determine whether there is a collision conflict from the two dimensions of time and distance. If a conflict occurs, record the obstacle conflict information, including the obstacle's lateral and longitudinal speeds and the time it arrives at the conflict point, as well as the ego vehicle conflict information, including the time the ego vehicle arrives at the conflict point and the distance from the conflict point.
5. The plug-in hybrid vehicle safety and energy-saving decision control method according to claim 1, characterized in that: In step S1, game decision theory is used to make decisions on the safety behavior of the vehicle, specifically: 1) When the vehicle is moving straight ahead and a dangerous obstacle is moving in front or on the left or right side and arrives at the predicted conflict point a certain time earlier than the vehicle, the vehicle does not accelerate; 2) When the vehicle is traveling straight ahead and a dangerous obstacle is traveling in front or on the left or right side, and reaches the predicted conflict point at the same time as the vehicle within a certain time range, and there is a risk of collision, the vehicle will slow down; 3) When the vehicle is moving straight ahead, dangerous obstacles are moving in front or on the left or right sides, and there is no conflict with the vehicle, so the vehicle moves normally; 4) When the ego vehicle turns, a dangerous obstacle is moving ahead and reaches the predicted conflict point at the same time as the ego vehicle within a certain time range, and there is a risk of collision, so the ego vehicle slows down.
6. The plug-in hybrid vehicle safety and energy-saving decision control method according to claim 1, characterized in that: The step S2 is specifically as follows: The prediction of dangerous obstacle trajectories, trajectory curvature restrictions, and safety behavior decision results are virtualized as control targets, and a discrete vehicle longitudinal dynamics model is constructed. With vehicle fuel economy, safety, and driving comfort as optimization targets, a multi-objective coordinated cost optimization function is constructed. Under system constraints, the vehicle's running trajectory and the optimal control algorithm are combined to perform longitudinal vehicle speed planning point by point, and the target object state is updated in real time. The local longitudinal safety speed planning is dynamically updated according to the actual vehicle speed. When the vehicle makes a safety behavior decision of deceleration or non-acceleration, the local longitudinal safe and economical speed is planned to intervene in the driving demand power in advance.
7. A plug-in hybrid vehicle safety and energy-saving decision control method according to claim 6, characterized in that: The vehicle longitudinal dynamics model in step S2 is mathematically expressed as: In the formula, x is the state variable, k represents the kth sampling time, represents the system matrix, y is the system output, C is the output matrix, u is the control input, with the vehicle speed or acceleration as the control variable, and w is the system disturbance, with the obstacle speed or acceleration as the disturbance variable.
8. The plug-in hybrid vehicle safety and energy-saving decision control method according to claim 1, characterized in that: In step S3, the model predictive control algorithm is used to perform longitudinal vehicle speed following control, specifically: Select the prediction time window T p = output y within [k+1:k+p] p The square of the difference between (i|k) and the target vehicle speed v(i) is accumulated to optimize the control variable sequence with the minimum objective function in the corresponding prediction time domain, and the first value in the sequence is used as the total required torque value of the current vehicle, and this process is repeated at the next moment; in min Oh, oh. max Where, the target speed v(i) is the longitudinal safety and economic speed reference curve of the vehicle, and y p (i|k) is the output obtained by real-time iteration of the vehicle dynamics model, u min 、u max Indicates the upper and lower limits of the current required torque.
9. A system based on the plug-in hybrid vehicle safety and energy-saving decision control method according to claim 1, characterized in that: The system includes: Environmental perception module, used to obtain environmental information around the target vehicle, including obstacle information, through intelligent sensors; The obstacle filtering module is used to filter the obstacles around the target vehicle to obtain dangerous obstacles; The trajectory prediction module is used to predict the trajectory of the dangerous obstacles and the ego vehicle obtained by the obstacle filtering module; A collision detection module is used to perform collision detection between the dangerous obstacles obtained by the obstacle filtering module and the vehicle; Safety behavior decision module, used to make safety behavior decisions for the ego vehicle based on trajectory prediction and collision detection results; The local longitudinal speed planning module is used to plan the vehicle speed based on the vehicle safety behavior decision results to obtain the longitudinal safe and economical vehicle speed; The speed following control module is used to perform longitudinal speed following control of the vehicle.
Citation Information
Patent Citations
Mathematical quantisation method for vehicle multiple target coordinating type self-adapting cruise control performance
CN101425106A
Vehicle trajectory optimization method based on vehicle network for fuel saving and driving comfort
CN111968377A
