Space-time multi-scale emergency priority passage trajectory planning method based on IVCPS

Through IVCPS technology, combined with Petri network and timing logic specifications, a priority pass scheme for emergency vehicles with multiple time and space scales was designed, which solved the problem of coordination between the traffic efficiency and safety of multiple emergency vehicles and social vehicles on the expressway, and achieved rapid and safe emergency vehicles Pass and minimize the effect on social vehicles.

CN119049275AInactive Publication Date: 2024-11-29CHONGQING UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411077079.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-07
Publication Date
2024-11-29
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing technology lacks a unified multi-time and spatial scale optimization framework and cross-scale coordination mechanism in the priority passage of highway emergency vehicles, making it difficult to effectively solve the problem of traffic efficiency and safety coordination between multiple emergency vehicles and social vehicles, and connect intelligent networks. Information acquisition and functional design in the environment are not in-depth enough.

Method used

Using the multi-scale emergency priority trajectory planning method based on IVCPS, the coordinated controllers of the vehicle layer and fleet layer are designed through the combination of Petri network and timing logic specifications to achieve efficient coordination between emergency vehicles and social vehicles, and build multi-space scales The collaborative optimization mechanism ensures rapid and safe passage of emergency vehicles, while minimizing the impact on social vehicle traffic efficiency.

Benefits of technology

While achieving rapid and safe passage of multiple emergency vehicles, it minimizes the impact on social vehicle traffic efficiency, providing solutions with theoretical significance and practical application value.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119049275A_ABST
    Figure CN119049275A_ABST
Patent Text Reader

Abstract

The invention relates to an IVCPS-based space-time multi-scale emergency priority passage trajectory planning method, and belongs to the field of intelligent traffic management and control. According to the method, a highway emergency vehicle priority passing scene is divided from multiple spatial-temporal scales to form three spatial-temporal scales including a vehicle layer, a motorcade layer and a road section layer; in a road section layer, a Petri network and linear sequential logic are combined, a priority passing problem is converted into an optimal accessibility problem, and the influence on social vehicle passing efficiency is minimized while priority passing of emergency vehicles is ensured; in a vehicle layer and a motorcade layer, an emergency priority multi-vehicle cooperative control method for a hybrid motorcade is provided, and efficient coordination of emergency vehicles and social vehicles is realized through establishment of a vehicle model and a motorcade model and a hierarchical control strategy based on a control barrier function.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of intelligent traffic management and control, and relates to a spatiotemporal multi-scale emergency priority passage trajectory planning method based on IVCPS. Background Art

[0002] The rapid and safe passage of emergency vehicles on highways is the key to carrying out emergency rescue work. When a large-scale sudden emergency occurs, highways can provide fast, smooth and safe transportation channels for emergency rescue vehicles, and facilitate the rapid passage of emergency rescue vehicles. When multiple emergency vehicles and social vehicles are mixed on the highway, there is competition for road resources between different vehicles. In addition, the high speed, high density and high variability of highway traffic flow make the priority passage of emergency vehicles more complicated and challenging. Therefore, it is necessary to design a systematic priority passage solution for sections with multiple emergency vehicles to improve traffic efficiency and driving safety.

[0003] With the development of computer, communication, Internet, automation and other technologies, IVCPS (Intelligent Vehicles Cyber ​​Physical Systems) information-physical system architecture provides a new idea for solving the problem of multiple emergency vehicle traffic. IVCPS refers to the integration of various information technologies and physical systems to achieve comprehensive perception integration, data sharing, collaborative decision-making and control of vehicles and road environments, thereby meeting the growing demand for intelligent traffic management. In the context of IVCPS, physical entities such as smart cars are integrated with information data through perception, communication and other technologies, which can realize collaborative interactive control between vehicles, thereby achieving reasonable planning of emergency vehicles.

[0004] The existing research on emergency vehicle priority planning strategies still has the following deficiencies:

[0005] (1) The priority passage of emergency vehicles on highways involves multiple spatiotemporal scales, such as vehicles, fleets, and road sections. Traffic participants at different spatiotemporal scales have significant differences and complex interactive coupling relationships in terms of behavioral logic, optimization objectives, and constraints. Most existing studies are based on a single spatiotemporal scale and lack a unified multi-spatiotemporal scale optimization framework and cross-scale coordination mechanism, making it difficult to effectively solve the problems of mixed fleet coordination control and group goal optimization.

[0006] (2) In highway emergency scenarios, multiple emergency vehicles are randomly distributed in different mixed fleets and travel together with social vehicles. Most existing studies focus only on a single optimization objective, such as emergency vehicle travel efficiency or travel time, while paying less attention to the impact of social vehicles. Traditional single-vehicle control methods are difficult to effectively coordinate the interactions between vehicles in a mixed fleet.

[0007] (3) Existing research is not deep enough in the acquisition and fusion application of full-dimensional information in the intelligent network environment, the functional positioning and interaction mechanism design of the information space and physical space in the intelligent network environment are not clear enough, and a hierarchical and systematic emergency priority passage plan has not yet been formed. Existing research on IVCPS enabling emergency vehicles to have priority passage still has a lot of room for improvement. Summary of the invention

[0008] In view of this, the purpose of the present invention is to provide a spatiotemporal multi-scale emergency priority passage trajectory planning method based on IVCPS, which ensures the rapid and safe passage of multiple emergency vehicles through multi-vehicle collaborative control and multi-target trajectories while minimizing the impact on the traffic efficiency of social vehicles.

[0009] In order to achieve the above object, the present invention provides the following technical solutions:

[0010] A spatiotemporal multi-scale emergency priority passage trajectory planning method based on IVCPS, which divides the highway emergency vehicle priority passage scenario from multiple spatiotemporal scales to form three spatiotemporal scales: vehicle layer, fleet layer and section layer; at the section layer, a global priority passage strategy for emergency vehicles is planned, and formation optimization at the fleet layer and motion control at the vehicle layer are guided; at the vehicle layer and fleet layer, real-time status information is collected and transmitted to the section layer to provide a basis for strategy generation at the section layer.

[0011] This method uses a combination of Petri nets and temporal logic specifications to solve the emergency vehicle priority trajectory:

[0012] Firstly, the STL (Signal Temporal Logic) constraints are transformed into CBF (Control Barrier Function) constraints, and the vehicle-level controller and fleet-level controller are designed respectively by combining the dynamic model and CBF constraints.

[0013] Design a vehicle-fleet hierarchical control weight allocation mechanism that coordinates the vehicle layer and the fleet layer: By introducing a weight coefficient in the vehicle layer controller, the weight coefficient is the weight coefficient of the CBF constraint issued by the fleet layer to the vehicle layer, which is used to adjust the control right allocation between the vehicle layer and the fleet layer controller, thereby achieving dynamic coordination of vehicle layer and fleet layer control instructions to adapt to different task requirements and scenarios;

[0014] Construct a highway colored Petri net model to reflect the movement and lane-changing behavior of vehicles on the highway; combine the highway colored Petri net model and linear temporal logic specifications to construct a combined Petri net that uniformly describes vehicle behavior and emergency vehicle priority rules, thereby transforming the emergency vehicle priority trajectory planning problem into the optimal reachability problem of the combined Petri net; construct a MILP model to solve the optimal reachability problem of the combined Petri net to obtain the optimal execution sequence, transform the optimal execution sequence into an execution sequence suitable for the highway colored Petri net model, and obtain the emergency vehicle priority trajectory by optimizing the execution sequence;

[0015] The priority trajectory is executed through collaborative optimization of three spatiotemporal scales: vehicle level, fleet level, and road section level.

[0016] Based on the unique multi-entity heterogeneity and spatiotemporal scale diversity of IVCPS, this invention proposes a highway emergency vehicle priority passage scheme based on multi-spatiotemporal scale collaboration, and combines multi-spatiotemporal scale analysis methods to decompose the problem into sub-problems at three spatiotemporal scales: vehicle layer, fleet layer, and road section layer. At the vehicle layer and fleet layer, an emergency priority multi-vehicle collaborative control method for mixed fleets is proposed. Through the establishment of vehicle models and fleet models, as well as a hierarchical control strategy based on control obstacle functions, efficient coordination between emergency vehicles and social vehicles is achieved. At the road section layer, a multi-objective trajectory planning method for emergency vehicle priority passage is proposed. Combining Petri nets and linear temporal logic, the priority passage problem is transformed into an optimal accessibility problem, which ensures the priority passage of emergency vehicles while minimizing the impact on the efficiency of social vehicles.

[0017] The beneficial effects of the present invention are that the present invention ensures the rapid and safe passage of multiple emergency vehicles through multi-vehicle collaborative control and multi-target trajectories while minimizing the impact on the traffic efficiency of social vehicles, and has important theoretical significance and practical application value.

[0018] Other advantages, objectives and features of the present invention will be described in the following description to some extent, and to some extent, will be obvious to those skilled in the art based on the following examination and study, or can be taught from the practice of the present invention. The objectives and other advantages of the present invention can be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below in conjunction with the accompanying drawings, wherein:

[0020] Figure 1 Schematic diagram of the emergency vehicle priority framework based on multiple spatiotemporal scales;

[0021] Figure 2 It is a schematic diagram of the multi-temporal and spatial scale collaborative architecture based on Petri nets;

[0022] Figure 3 This is a diagram of the hierarchical controller architecture;

[0023] Figure 4 This is a schematic diagram of a colored Petri net model;

[0024] Figure 5 Schematic diagram of the multi-temporal and spatial scale execution framework of priority access trajectories. DETAILED DESCRIPTION

[0025] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner, and the following embodiments and features in the embodiments can be combined with each other without conflict.

[0026] Aiming at solving the problem of balancing the traffic efficiency of multiple emergency vehicles and social vehicles in highway scenarios, the present invention proposes an emergency vehicle priority passage scheme based on IVCPS multi-temporal and spatial scales. Through multi-vehicle collaborative control and multi-target trajectories, the rapid and safe passage of multiple emergency vehicles can be ensured, while the impact on the traffic efficiency of social vehicles can be minimized.

[0027] In one embodiment of the present invention, based on the unique multi-entity heterogeneity, spatiotemporal scale diversity, and deep integration of information and physics of IVCPS, a multi-spatiotemporal scale analysis method is adopted to divide the priority passage scenario of emergency vehicles on highways from three multi-spatiotemporal scales: vehicle layer, fleet layer, and road section layer. Figure 1 As shown in the figure. The vehicle layer, fleet layer and section layer build a multi-level collaborative interconnection in the physical space and information space. On the one hand, the global priority traffic strategy at the section layer guides the formation optimization at the fleet layer and the motion control at the vehicle layer through cross-level information transmission and task decomposition; on the other hand, the real-time status information at the vehicle layer and fleet layer provides a basis for the strategy generation at the section layer through cross-level perception and fusion. Through this bottom-up status feedback and top-down strategy guidance, a closed-loop multi-layer collaborative optimization mechanism is built, realizing dynamic collaboration and intelligent decision-making from local to global.

[0028] This embodiment uses a method combining Petri nets and temporal logic specifications to achieve the description and solution of the emergency vehicle priority passage problem on multiple time and space scales, such as Figure 2 Considering the information-physical fusion characteristics of IVCPS, Petri nets are used to map physical entities of different scales into information space for modeling, and temporal logic specifications are used to characterize task requirements and constraints of different scales.

[0029] At the vehicle layer and fleet layer, Petri nets are used to model the behaviors of vehicles and fleets respectively. At the same time, information temporal logic specifications are introduced to define the real-time control tasks and safety constraints that vehicles and fleets need to meet. At the road section layer, Petri nets are used to model the traffic flow of the road section to describe the priority strategy and safety properties. The combination of Petri nets and temporal logic enhances the expression ability of the model at different time and space scales.

[0030] In this embodiment, the coordinated control strategy for emergency priority passage at the vehicle level and the fleet level is designed as follows:

[0031] 1. Design of vehicle layer controller

[0032] Different objective functions and weight coefficients are designed for emergency vehicles and social vehicles, and vehicle priority attributes are introduced in the optimization objectives. By giving emergency vehicles higher tracking performance weights and smaller control cost weights, priority guarantee for emergency vehicle trajectory tracking is achieved. The design process of the controller is as follows:

[0033] (1) Vehicle layer controller input and output

[0034] The input of the vehicle-level controller includes: vehicle status information (position, speed, acceleration, etc.), local environment perception information (relative position and speed with surrounding vehicles, etc.), vehicle type information (emergency vehicle or civilian vehicle), and reference trajectory issued by the upper-level fleet-level controller. Its output is the optimal acceleration control command that meets various constraints.

[0035] (2) Objective function design

[0036] The objective function of the vehicle layer controller consists of two parts: the trajectory tracking error term and the control amount change rate term, which are used to measure the trajectory tracking performance and the stability of the control command respectively. Considering the emergency priority characteristics, different weight coefficients are used for emergency vehicles and social vehicles in the trajectory tracking error term. The mathematical description of the objective function is as follows:

[0037]

[0038]

[0039] Among them, x and x refare the vehicle state and reference trajectory respectively, It is the derivative of the control quantity u, that is, the acceleration command. i is the trajectory tracking error weight matrix, which is Q1 and Q2 for emergency vehicles and social vehicles respectively, and Q1>Q2, indicating that the trajectory tracking performance of emergency vehicles has a higher priority. R is the weight matrix of the control variable change rate, and its value reflects the importance attached to the stability of the control command. f , t0 represents the upper and lower limits of the integration interval.

[0040] (3) CBF (Control Barrier Function) representation of STL (Signal Temporal Logic) constraints

[0041] In order to unify the various constraints faced by the vehicle layer (including safety constraints, feasibility constraints and STL task constraints, etc.) into the form of CBF constraints, it is first necessary to convert the STL constraints into CBF representation.

[0042] Definition: Given an STL formula If there exists a continuously differentiable function If and only if When true, Then it is called for The control barrier function.

[0043] Based on the above definition, common STL constraints can be converted into CBF:

[0044] ①Speed ​​constraint: The CBF of this constraint is expressed as:

[0045]

[0046] Among them, G is a constant coefficient matrix, v(t) represents the vehicle speed, v max represents the maximum speed of the vehicle, and x(t) represents the vehicle state function.

[0047] ②Interval constraints between vehicles: The CBF of this constraint is expressed as:

[0048]

[0049] Among them, p i (t) represents the position of vehicle i, p j (t) represents the position of vehicle j, d safe Indicates safe vehicle spacing, x i (t) represents the speed of vehicle i, x j(t) represents the speed of vehicle j.

[0050] After converting all STL constraints faced by the vehicle layer into CBF, a set of CBF constraints can be obtained: Among them, N c is the total number of constraints.

[0051] (4) Design of vehicle-level controller based on CBF

[0052] Combining the vehicle dynamics model and CBF constraints, the vehicle-level controller design problem can be formulated as a quadratic programming (QP) problem with CBF constraints:

[0053]

[0054] Among them, f(x) and g(x) are the drift term and control term of the vehicle dynamics model, and for The Lie derivatives of the state x and control u, α(·), are K-like functions. represents the derivative of the vehicle dynamics model x, u min Indicates the minimum acceleration control amount, u max Indicates the maximum acceleration control amount.

[0055] The objective function of the QP problem is the sum of the squared acceleration term and the tracking error term shown in equation (1). The constraints include the vehicle dynamics equation, all CBF constraints, and the upper and lower bounds of the control quantity. Solving the QP problem can obtain the optimal acceleration command that satisfies various constraints. The effective set method is used to solve the above QP problem. By actively setting constraints, the original problem is transformed into an equivalent unconstrained optimization problem, and then the gradient descent method is used for rapid solution.

[0056] 2. Design of Fleet-Level Controller

[0057] In the design of the fleet-level controller, the emergency priority factor is introduced on the basis of formation control and trajectory optimization. Compared with the existing methods, the priority distinction between emergency fleets and ordinary fleets is introduced in the optimization objectives. By giving the emergency fleet a higher formation weight and trajectory weight, the priority of the emergency fleet's formation behavior and motion trajectory can be achieved.

[0058] (1) Fleet level controller input and output

[0059] The input of the fleet controller includes: the status information of the vehicles under its jurisdiction, the fleet mission requirements (such as fleet formation, fleet splitting, etc.) and the emergency priority instructions issued at the road section level. Its output is the reference trajectory information of each vehicle.

[0060] (2) Objective function design

[0061] The objective function of the fleet-level controller includes fleet formation performance and emergency priority performance, which can be mathematically described as follows:

[0062]

[0063] Among them, N f is the total number of fleets, N i is the set of vehicle i’s platoon neighbors, d ij is the expected platoon spacing of vehicles, N e is the number of emergency vehicle fleets, is the reference speed of the emergency convoy, N s is the number of social vehicle fleets, is the reference speed of the social convoy. In formula (6), the first term in the integral symbol is the convoy formation performance, which is expressed by the relationship between all vehicles and their expected convoy positions d ij The second and third items are the speed deviations of the emergency team and the social team, respectively. and is the reference speed of the two types of fleets, usually To reflect the priority of emergency convoys; w1~w3 are the weight coefficients of each item. In order to highlight the priority of emergency, w2>w3 is usually taken.

[0064] (3) CBF Representation of STL Constraints

[0065] Similar to the vehicle layer, the fleet layer controller needs to meet multiple constraints, including fleet formation constraints, fleet trajectory constraints, and STL task constraints issued by the high-level layer. In order to unify the STL constraints with other constraints, the CBF method is also used to convert them.

[0066] For example, for the STL constraint of a convoy maintaining its formation:

[0067]

[0068] where v f is the set of vehicles in fleet f, G f (·) represents a constant coefficient matrix. This constraint can be transformed into the following CBF:

[0069]

[0070] After converting all STL constraints faced by the fleet layer into CBF, a set of CBF constraints can be obtained:

[0071] (4) Design of fleet-level controller based on CBF

[0072] Combining the fleet dynamics model and CBF constraints, the fleet-level controller design problem can be formulated as a nonlinear programming (NLP) problem with CBF constraints:

[0073]

[0074] Among them, p i and v i are the position and speed of vehicle i, For p i The derivative of The road is drivable area.

[0075] The objective function of the NLP problem is the sum of the formation performance, emergency priority performance and fuel economy as shown in equation (6). The constraints include the vehicle kinematic equations, all CBF constraints and the road drivable area constraints. Solving the NLP problem can obtain the optimal reference trajectory of the fleet that meets all kinds of constraints.

[0076] 3. Design of vehicle-fleet hierarchical controller coordination mechanism

[0077] Based on the vehicle-level and fleet-level controllers designed above, a hierarchical control weight allocation mechanism is proposed. By introducing weight coefficients, dynamic coordination of fleet-level and vehicle-level control commands is achieved. This mechanism can adjust the control rights allocation of the two-layer controllers in real time according to factors such as task type and traffic environment, making the hierarchical control system more adaptable.

[0078] (1) Task allocation and decomposition

[0079] In the hierarchical control framework, such as Figure 3 As shown in Figure 1, the fleet controller is responsible for receiving high-level task instructions from the road segment level and decomposing them into a series of STL constraints applicable to the fleet. These constraints cover multiple aspects such as formation maintenance, speed control, trajectory tracking, etc., forming the task objectives of the fleet.

[0080] After solving the NLP problem shown in equation (8), the fleet controller further decomposes the optimized fleet reference trajectory into reference trajectories of each vehicle and sends it to the vehicle controller. At the same time, the fleet converts the STL constraint into a CBF constraint and sends it to the vehicle controller as one of the constraints for the vehicle controller to solve the QP problem.

[0081] (2) Control weight adjustment

[0082] In the vehicle-level optimal control problem shown in equation (9), a weight coefficient λ∈[0,1] is introduced to assign weights to the CBF constraints issued by the fleet level, namely:

[0083]

[0084] Among them, λ is the weight coefficient of the fleet-level CBF constraint in the vehicle-level optimization control. The first constraint is the original CBF constraint at the vehicle level, and the second constraint is the weighted CBF constraint issued by the fleet level.

[0085] Based on the integrated characteristics of IVCPS perception, communication and other technologies and information data fusion, the specific value of λ can be dynamically adjusted according to factors such as the mission type of the fleet and the traffic environment in which it is located: by introducing a control weight adjustment mechanism, the flexibility and robustness of the hierarchical control system can be further enhanced, enabling it to adapt to different mission requirements and application scenarios.

[0086] In this embodiment, the global trajectory planning for emergency priority passage at the section level is as follows:

[0087] 1. Section-level Petri net modeling for emergency priority

[0088] In order to describe the scenario of emergency vehicles having priority, a highway colored Petri net model HWCPN for emergency priority is defined, and a highway colored Petri net model HWCPN=(P,T,A,∑,V,C,G,E,I) is constructed. Figure 4 As shown in the figure, it is a highway colored Petri net model constructed in this embodiment. The model consists of 9 libraries (such as Figure 4 The oval in the middle) and 14 transitions (such as Figure 4 The model can reflect the movement and lane-changing behavior of vehicles on a three-lane highway.

[0089] Specifically, the place set P represents cells on the highway, and each place corresponds to a cell. The location of each place is represented by a two-dimensional coordinate (x, y), where x represents the longitudinal position and y represents the lane number. For example, place p(3,2) represents the cell in the 3rd column and the 2nd lane.

[0090] The transition set T represents the movement and lane change behavior of the vehicle, and defines two types of transitions:

[0091] ① Longitudinal movement transition T f [i,j]: Indicates that the vehicle moves from cell p(i,j) to the next cell p(i+1,j), staying in the same lane.

[0092] ② Oblique lane change T lc [i,j,k]: indicates that the vehicle moves diagonally from cell p(i,j) to cell p(i+1,k) and completes the lane change. k=j-1 indicates a lane change to the right, and k=j+1 indicates a lane change to the left.

[0093] The directed arc set A represents the causal relationship between vehicle movement and lane change. For each longitudinal movement transition T f [i,j], add a line from p(i,j) to T f [i,j] input arc, and an arc from T f The output arc from [i,j] to p(i+1,j). For each diagonal lane change transition T lc [i,j,k], add a line from p(i,j) to T lc [i,j,k] input arc, and an arc from T lc The outgoing arc from [i,j,k] to p(i+1,k).

[0094] Next, linear temporal logic (LTL) is used to formally describe the emergency vehicle priority rules on highway sections, including lane selection rules, vehicle avoidance rules, speed control rules, and safety rules. Combining the HWCPN model and LTL rules, a Composed PetriNet (Co-PN) is constructed to uniformly describe vehicle behavior and emergency vehicle priority rules.

[0095] A combinatorial Petri net is a nine-tuple Co-PN = (P, T, F, ∑, L, φ, M0, R, λ), where:

[0096] (1) (P, T, F) is a basic Petri net structure. Represents a flow relationship;

[0097] (2)∑ is a finite alphabet representing the set of atomic propositions;

[0098] (3)L:P→2 ∑ is the place labeling function, mapping each place to a subset of ∑;

[0099] (4)φ is the LTL formula defined on ∑, representing the priority rule;

[0100] (5) M0:P→N is the initial identification;

[0101] (6) is the Büchi repository, corresponding to the Büchi automaton B φ The state set Q of

[0102] (7)λ:T→Q×Q is the transition labeling function, which maps each transition to a pair of Büchi states, indicating the state transition caused by the occurrence of the transition.

[0103] 2. Mixed integer linear programming of emergency priority trajectory at the section level

[0104] On this basis, the emergency vehicle priority trajectory planning problem is transformed into finding a minimum cost execution path from the initial identifier to the target identifier in the combinatorial Petri net state space (optimal reachability problem). The optimal reachability problem of Co-PN can be systematically transformed into a MILP problem by introducing decision variables and linear constraints.

[0105] The construction method of the mixed integer linear programming (MILP) optimization model is to transform the multi-objective optimization problem into a single-objective problem by using the weighted summation method according to the actual demand for emergency vehicle priority, and control the relative importance of different objectives by adjusting the weight coefficients of different objectives. The construction method of the optimization objective function and constraint conditions in the MILP model is as follows:

[0106] (1) Construction of optimization objective function

[0107] Define the following decision variables and objective function:

[0108] ① Emergency vehicle travel time. Introduce 0-1 decision variable x t ∈[0,1] represents the occurrence of transition t. The transition sequence experienced by the emergency vehicle is denoted as T E , the corresponding travel time is τ t , then the emergency vehicle travel time can be expressed as:

[0109]

[0110] ② The average speed of social vehicles decreases. Introduce 0-1 decision variable y v ∈[0,1] represents the deceleration behavior of vehicle v. The speeds of the kth social vehicle before and after the implementation of priority passage are v k and v′ k , then the average velocity attenuation can be expressed as:

[0111]

[0112] Here, K represents the number of modes.

[0113] ③The number of lane changes of social vehicles. Introduce 0-1 decision variable z v ∈[0,1] represents the lane-changing behavior of vehicle v, and the number of lane changes of social vehicles can be expressed as:

[0114]

[0115] Where V S Represents the set of all social vehicles.

[0116] Combining the above three objective functions, the MILP objective function of the emergency vehicle priority problem can be obtained:

[0117] min{W1f1+W2f2+W3f3}(13)

[0118] Among them, W1, W2, and W3 are weight coefficients of each optimization objective, which can be set according to actual needs.

[0119] (2) Construction of constraint conditions

[0120] In order to ensure that the MILP model can reasonably describe the emergency vehicle priority process, the following constraints need to be constructed:

[0121] ① Flow conservation constraint. For any place p∈P, an integer variable m is introduced p ∈Z represents its identity, and constructs the constraints:

[0122]

[0123] Among them, ·p represents the place before Co-PN execution, and p· represents the place after Co-PN execution.

[0124] This constraint ensures the conservation of the number of place identifiers before and after the Co-PN execution.

[0125] ② Transition constraints. For any transition t∈T, construct constraints:

[0126]

[0127] Where ·t represents the input place set of transition t.

[0128] This constraint ensures that transition t can only occur when all input places of transition t have sufficient identifiers, which complies with the occurrence rules of Petri nets.

[0129] ③ Mutually exclusive transition constraints. For any pair of mutually exclusive transitions t1, t2∈T, construct the constraint:

[0130]

[0131] This constraint avoids multiple conflicting behaviors of the same vehicle at the same time.

[0132] ④ Vehicle behavior constraints. Based on the three-lane highway scenario and conventional traffic rules, linear constraints on vehicle overtaking, lane changing, etc. are constructed, for example:

[0133] For any vehicle v, its position p on road segment r v satisfy:

[0134]

[0135] in represents the set of motion transitions of vehicle v on road segment r, lt Represents the displacement length of transition t.

[0136] For any two vehicles v1 and v2, construct a safe distance constraint to avoid collision:

[0137]

[0138] For the transition t corresponding to the lane-changing behavior, construct the constraint: It indicates that the occurrence of transition t implies the lane-changing behavior of vehicle v.

[0139] ⑤Objective function related constraints: According to the corresponding relationship between each optimization objective and the Co-PN model, the corresponding linear constraints are constructed.

[0140] Through the construction of the above objective function and constraints, a complete MILP optimization model for emergency vehicle priority can be obtained.

[0141] (3) MILP model solution

[0142] The commercial optimization solver Gurobi 9.1 is used to build and solve the MILP model. The Gurobi solver provides an easy-to-use Python programming interface that can easily call the Gurobi optimization engine to solve the model. Based on Gurobi, an automated MILP model generation and optimization solution process can be implemented: first, according to the definition of the Co-PN model, the model variables, objective functions, and constraints are automatically extracted and converted into mathematical forms supported by Gurobi; then, the Gurobi engine is called to solve the MILP model, and the optimization process and key performance indicators are recorded during the solution process; finally, the solution results are fed back to the Co-PN model to output the optimized vehicle trajectory strategy.

[0143] By solving the MILP model, we can obtain an optimal execution sequence σ from the initial mark to the target mark. However, the transition sequence in σ is based on the Co-PN model and needs to be further converted into an execution sequence on the HWCPN before it can be used as the priority passage trajectory for emergency vehicles.

[0144] Let π denote the optimized HWCPN execution sequence, then π can be expressed as:

[0145] π=t1t2…t n ,t i ∈T (19)

[0146] Where T is the set of transitions of HWCPN. The sequence π gives a priority passage trajectory for emergency vehicles, that is, by triggering each transition according to this sequence, the emergency vehicle can reach the destination as soon as possible while meeting the passage regulations.

[0147] 3. Execution mechanism based on emergency priority passage trajectories at multiple spatiotemporal scales

[0148] By further refining and optimizing the MILP solution, a refined HWCPN priority passage trajectory π* is obtained. In order to implement this trajectory in the actual highway traffic system and realize the priority passage of emergency vehicles, a trajectory execution mechanism based on multi-time and space scale division is proposed, such as Figure 5 shown.

[0149] According to the three divided time and space scales, the execution of the priority trajectory π* needs to be collaboratively optimized at the vehicle level, fleet level and road section level. Combining the concepts and characteristics of IVCPS such as sharing, collaboration and openness, the collaborative optimization includes:

[0150] At the vehicle level, emergency vehicles and social vehicles perceive their own status and surrounding environment in real time through on-board sensors, and use vehicle-to-vehicle communication to exchange information and coordinate behaviors. Emergency vehicles quickly plan motion parameters according to the requirements of the priority trajectory π*, and social vehicles adjust speed or change lanes in time after receiving the avoidance request to make way for emergency vehicles.

[0151] At the fleet level, emergency vehicles and social vehicles are located in the fleet according to the priority trajectory π * Optimize formation and dispatch. Teams negotiate lane change and overtaking strategies through V2V communication to ensure priority for emergency vehicles. At the same time, the team layer also interacts with the edge cloud platform through vehicle-cloud collaboration to obtain traffic information in a wider range and optimize the overall driving strategy of the team.

[0152] At the road section level, the roadside infrastructure perceives the traffic status of the road section in real time through vehicle-to-infrastructure collaboration (V2I) and uploads the data to the edge / regional cloud platform. The cloud platform analyzes the traffic situation within the road section and plans the optimal passage path for emergency vehicles at the road section level. When the actual traffic status deviates significantly, the cloud platform can also trigger the priority passage trajectory π * re-planning.

[0153] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution, which should be included in the scope of the claims of the present invention.

Claims

1. A spatiotemporal multi-scale emergency priority passage trajectory planning method based on IVCPS, Features: The method divides the highway emergency vehicle priority passage scenario from multiple spatiotemporal scales to form three spatiotemporal scales: vehicle layer, fleet layer and section layer. At the section layer, the global priority passage trajectory of the emergency vehicle is planned, and the formation optimization of the fleet layer and the motion control of the vehicle layer are guided. At the vehicle layer and fleet layer, real-time status information is collected and transmitted to the section layer to provide a basis for the trajectory generation at the section layer. This method uses a combination of Petri nets and temporal logic specifications to solve the emergency vehicle priority trajectory: Firstly, the STL (Signal Temporal Logic) constraints are transformed into CBF (Control Barrier Function) constraints, and the vehicle-level controller and fleet-level controller are designed respectively by combining the dynamics model and CBF constraints. Design a vehicle-fleet hierarchical control weight allocation mechanism that coordinates the vehicle layer and the fleet layer: By introducing a weight coefficient in the vehicle layer controller, the weight coefficient is the weight coefficient of the CBF constraint issued by the fleet layer to the vehicle layer, which is used to adjust the control right allocation between the vehicle layer and the fleet layer controller, thereby achieving dynamic coordination of vehicle layer and fleet layer control instructions to adapt to different task requirements and scenarios; Construct a highway colored Petri net model to reflect the movement and lane-changing behavior of vehicles on the highway; combine the highway colored Petri net model and linear temporal logic specifications to construct a combined Petri net that uniformly describes vehicle behavior and emergency vehicle priority rules, thereby transforming the emergency vehicle priority trajectory planning problem into the optimal reachability problem of the combined Petri net; construct a MILP model to solve the optimal reachability problem of the combined Petri net to obtain the optimal execution sequence, transform the optimal execution sequence into an execution sequence suitable for the highway colored Petri net model, and obtain the emergency vehicle priority trajectory by optimizing the execution sequence; The priority trajectory is executed through collaborative optimization of three spatiotemporal scales: vehicle level, fleet level, and road section level.

2. The method according to claim 1, Features: The design process of the vehicle layer controller includes: Design the input and output of the vehicle-level controller: the input includes the vehicle's own state information, local environment perception information, vehicle type information and the reference trajectory issued by the fleet-level controller; the output includes the optimal acceleration control command that meets the constraints; Design the objective function of the vehicle layer controller: In the formula, x and x ref Represent the vehicle state and reference trajectory respectively, Represents the derivative of the control quantity u, that is, the acceleration command; Q i represents the trajectory tracking error weight matrix, Q 1 >Q 2 , indicating that the trajectory tracking performance of emergency vehicles has a higher priority; R represents the weight matrix of the control variable change rate, reflecting the importance attached to the stability of the control command; t f ,t 0 Respectively represent the upper and lower limits of the integral interval; The STL constraints of the vehicle layer are converted into CBF constraints. Combined with the vehicle dynamics model, the vehicle layer controller design is expressed as a quadratic programming problem with CBF constraints: Where f(x) and g(x) represent the drift term and control term of the vehicle dynamics model, respectively. represents the derivative of the vehicle dynamics model x, represents the CBF constraint, and express The Lie derivatives of the state x and control u, α(·) represents a K-like function, u min Indicates the minimum acceleration control amount, u max Indicates the maximum acceleration control amount.

3. The method according to claim 1, Features: The design process of the fleet-level controller includes: Design the input and output of the fleet controller: the input includes the status information of the fleet’s vehicles, the fleet’s mission requirements, and the emergency priority instructions issued by the road section; the output is the reference trajectory information of each vehicle; Design the objective function of the fleet-level controller: Where N f represents the total number of fleets, N i represents the set of vehicle i’s platoon neighbors, d ij represents the expected platoon spacing of vehicles, N e represents the number of emergency vehicle fleets, represents the reference speed of the emergency convoy, N s represents the number of social vehicle fleets, represents the social fleet reference speed, p i and p j denote the positions of vehicle i and vehicle j respectively, v i represents the speed of the vehicle, w 1 、w 2 、w 3 Represents the weight coefficient, set w 2 >w 3 , improve the priority of emergency fleet; The STL constraints at the fleet level are converted into CBF constraints. Combining the fleet dynamics model and CBF constraints, the fleet-level controller design is expressed as a nonlinear programming problem with CBF constraints: In the formula, For p i The derivative of represents the CBF constraint, Indicates the drivable area of ​​the road.

4. The method according to claim 1, Features: Designing the vehicle-team hierarchical control weight allocation mechanism includes: The fleet-level controller receives the task instructions of the road section level and decomposes them into STL constraints applicable to the fleet; the fleet-level controller solves the nonlinear programming problem according to the converted CBF constraints to obtain the fleet reference trajectory, decomposes it into reference trajectories of the vehicles under the fleet, and sends it to the vehicle-level controller; at the same time, the CBF constraints of the fleet level are sent to the vehicle level as one of the constraints for the vehicle-level controller to solve the quadratic programming problem; Assign a weight λ to the CBF constraints sent from the fleet layer to the vehicle layer, and change the vehicle layer controller to:

5. The method according to claim 1, Features: The linear temporal logic specification is a specification for describing the priority passage of emergency vehicles on highways, including lane selection specifications, vehicle avoidance specifications, speed control specifications and safety specifications.

6. The method according to claim 1, Features: Constructing the MILP model includes: 1) Construct the objective function, including the emergency vehicle travel time, the average speed decay of social vehicles and the number of lane changes of social vehicles; The emergency vehicle travel time is expressed as: Where t represents the transition in the highway colored Petrified network model, T E represents the sequence of transitions experienced by the emergency vehicle, τ t Indicates the travel time; x t ∈[0,1] is a decision variable, indicating the occurrence of transition t; The average speed decay of social vehicles is expressed as: In the formula, K represents the number of modes, v k and v′ k They represent the speed of the kth social vehicle before and after the implementation of priority passage; y v ∈[0,1] is the decision variable, which represents the deceleration behavior of vehicle v; The number of lane changes of social vehicles is expressed as: Where V S represents the set of all social vehicles; z v ∈[0,1] is the decision variable, which represents the lane-changing behavior of vehicle v; f 1 、f 2 、f 3 The objective function is obtained by weighted summation: min{w 1 f 1 +w 2 f 2 +w 3 f 3 } Where W 1 , W 2 , W 3 is the weight coefficient; 2) Construct constraints, including flow conservation constraints, transition occurrence constraints, transition mutual exclusion constraints, vehicle behavior constraints, and objective function related constraints.

7. The method according to claim 1, Features: The execution of the priority passage trajectory through the coordinated optimization of three spatiotemporal scales of vehicle level, fleet level and road section level includes: At the vehicle level, emergency vehicles and social vehicles use vehicle-to-vehicle communication to exchange information and coordinate behaviors. Emergency vehicles follow the priority trajectory π * The motion parameters are planned according to the requirements. After receiving the avoidance request, the private vehicle adjusts the speed or changes the lane in time to make way for the emergency vehicle. At the fleet level, emergency vehicles and social vehicles are located in the fleet according to the priority trajectory π * Optimize formation and dispatch. Use V2V communication between fleets to negotiate lane change and overtaking strategies to ensure the priority of emergency vehicles. At the same time, the fleet layer also interacts with the edge cloud platform through vehicle-cloud collaboration to obtain traffic information in a wider range and optimize the overall driving strategy of the fleet. At the road section level, the roadside infrastructure perceives the traffic status of the road section in real time through vehicle-road collaboration and uploads the data to the edge / regional cloud platform. The cloud platform analyzes the traffic situation within the road section and plans the optimal route for emergency vehicles at the road section level. When the actual traffic status deviates significantly, the cloud platform replans the priority route π * .