Observation-based cooperative control method for multiple intelligent vehicles with unknown input delays

By building intelligent vehicle kinematics model and multi-intelligent vehicle system model, using output feedback observer and adaptive switching strategy to estimate unknown input delays online, the problems of communication delays and packet loss between vehicles are solved, and the high level of intelligent fleets is achieved Stability and security coordinated control.

CN115257732BActive Publication Date: 2024-11-26TONGJI UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210773293.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-07-01
Publication Date
2024-11-26
Estimated Expiration
2042-07-01

AI Technical Summary

Technical Problem

The existing multi-intelligent vehicle collaborative control method is susceptible to communication delays and packet loss during communication between vehicles, and fails to effectively consider sensor input delays, resulting in insufficient control stability and security.

Method used

By constructing intelligent vehicle kinematics models and multi-intelligent vehicle system models, using output feedback observers to obtain vehicle observations, and online estimate of unknown input delays of each vehicle through adaptive switching strategies, designing multi-intelligent vehicle collaborative controllers to achieve Vertical control avoids communication between vehicles and only relies on on-board sensors to obtain information about the front vehicle.

Benefits of technology

It effectively improves the security and stability of vertical collaborative control of intelligent fleets, has strong anti-interference ability, and avoids the impact of communication delay and packet loss.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115257732B_ABST
    Figure CN115257732B_ABST
Patent Text Reader

Abstract

The present invention relates to a multi-intelligent vehicle cooperative control method with unknown input delay based on observation, including: constructing a kinematic model of an intelligent vehicle with unknown input delay; establishing a multi-intelligent vehicle system model based on the kinematic model of the intelligent vehicle and the topological architecture of the multi-intelligent vehicle; obtaining the observation value of the intelligent vehicle through an output feedback observer, and inputting the observation value into the multi-intelligent vehicle cooperative controller; offline obtaining several types of offline parameters of the multi-intelligent vehicle cooperative controller, and online switching several types of offline parameters through the adaptive switching strategy of the multi-intelligent vehicle system model until the stabilizer of the multi-intelligent vehicle system model is stable, and obtaining the unknown input delay of each intelligent vehicle; inputting the unknown input delay into the multi-intelligent vehicle cooperative controller to realize the longitudinal control of the multi-intelligent vehicle. Compared with the prior art, the present invention has the advantages of strong anti-interference ability, strong stability, high safety, etc.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of intelligent driving, and in particular to an observation-based collaborative control method for multiple intelligent vehicles with unknown input delays. Background Art

[0002] With the advent of the intelligent era, both the industrial and military fields are inseparable from the collaborative division of labor of multiple intelligent agents. In addition, many fields such as cargo transportation and distributed power generation control have realized distributed collaborative production of multiple agents. However, fields such as distributed collaborative control of fleets in traffic are still in the exploratory stage. The collaborative control problem of multi-agent systems is still a hot topic in current research, and it has important theoretical guiding significance for industrial production, military applications and other fields.

[0003] The development of multi-agent system technology provides a new way for intelligent control of complex, large-scale and nonlinear systems. Using multi-agent system technology for control can decompose the original single, large and nonlinear system according to the requirements of the control application. Using multi-agent system technology for control can also divide multiple agents functionally to complete the control task of the original complex system.

[0004] The multi-agent system not only has the characteristics of information sharing, easy expansion, high reliability, good flexibility and strong real-time performance of ordinary distributed systems, but also the agents can solve complex control problems through mutual communication, coordination and integration.

[0005] The earliest automatic cooperative control of vehicles can be traced back to the automatic speed control of vehicles, such as cruise control, and the later developed adaptive cruise control system. The adaptive cruise control system is the first driver assistance control system to enter the market, which has an important impact on the control of traffic flow. In this mode, the vehicle uses the information of the on-board sensor to design the cruise controller. In traditional cruise control, the vehicle is always required to maintain a preset speed without considering the changes in the surrounding traffic environment. In the adaptive cruise control system, the control goal of the vehicle is to keep the desired safe distance from the previous vehicle as much as possible, and control the speed of the vehicle itself to keep the same speed as the previous vehicle according to the speed measurement results of the forward-looking sensor (usually millimeter-wave radar or infrared laser). However, cruise control and adaptive cruise control are control strategies designed for a single vehicle. With the development of vehicle-to-vehicle communication and vehicle-to-road communication technology, when adaptive cruise control and these communication technologies are combined to control and dispatch the vehicle, the vehicle control system becomes a cooperative adaptive cruise control system. In the cooperative control system, in addition to using its own sensors to measure relevant data, the vehicle can also receive relevant information about neighboring vehicles through communication. In the cooperative adaptive cruise control system, vehicles in the convoy drive safely with a tighter spacing between vehicles, which improves vehicle throughput and has stronger anti-interference ability. It mainly realizes communication between vehicles through wireless communication, which can effectively improve the traffic capacity of the road and the stability of vehicle control. At present, cooperative adaptive control is the focus of researchers in related research fields and has also attracted widespread attention in countries around the world.

[0006] Among the existing multi-intelligent vehicle collaborative control methods, most use vehicle communication to transmit position and speed signals. In-vehicle communication will cause communication delay and packet loss, and the communication network between vehicles is also vulnerable to attacks, resulting in loss of control. Among the multi-intelligent vehicle control methods that use sensors, few methods consider the impact of sensor input delay. Summary of the invention

[0007] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and to provide an observation-based multi-intelligent vehicle collaborative control method with unknown input delay, which has strong anti-interference ability, strong stability and high safety.

[0008] The purpose of the present invention can be achieved by the following technical solutions:

[0009] A method for cooperative control of multiple intelligent vehicles with unknown input delays based on observation, comprising:

[0010] Construct a kinematic model of an intelligent vehicle with unknown input delays;

[0011] Establish a multi-intelligent vehicle system model based on the intelligent vehicle kinematic model and the topological architecture of multiple intelligent vehicles;

[0012] Obtain observation values ​​of the intelligent vehicles through the output feedback observer, and input the observation values ​​into the multi-intelligent vehicle cooperative controller;

[0013] Offline acquisition of several types of offline parameters of the multi-intelligent vehicle cooperative controller, online switching of several types of offline parameters through the adaptive switching strategy of the multi-intelligent vehicle system model, until the stabilizer of the multi-intelligent vehicle system model is stable, and the unknown input delay of each intelligent vehicle is obtained;

[0014] The unknown input delay is input into the multi-intelligent vehicle cooperative controller to achieve longitudinal control of the multi-intelligent vehicles.

[0015] Furthermore, the expression of the intelligent vehicle kinematic model with unknown input delay is:

[0016]

[0017] Among them, q i (t) represents the position of intelligent vehicle i, p i (t) represents the speed of intelligent vehicle i, a i (t) represents the acceleration of intelligent vehicle i, η represents the time constant of lag, h i represents the unknown input delay of intelligent vehicle i, and i represents the time from the 1st vehicle to the Nth vehicle.

[0018] Furthermore, in the topological architecture of the multi-intelligent vehicle, each vehicle only uses the vehicle-mounted sensor to obtain the position information of the preceding vehicle, and the expression of the subsystem of the vehicle i and the preceding vehicle i-1 is:

[0019]

[0020] Among them, Δδ i (t) is the error between the actual distance and the expected distance between vehicle i and the preceding vehicle i-1, Δδ i (t) = q i (t)-q i-1 (t)-d0, d0 is the safe distance between vehicle i and the preceding vehicle i-1;

[0021] The expression of the multi-intelligent vehicle system model is:

[0022]

[0023] x i (t) = [Δδ i (t), Δp i (t), Δa i (t)]T

[0024] y i (t) = Δδ i (t)

[0025]

[0026]

[0027] C i =[0 0 1]

[0028] ρ i (t) = [0 0 -B i u i-1 (th i-1 )] T .

[0029] Furthermore, the formation goal of the multi-intelligent vehicle system model is specifically:

[0030]

[0031] Furthermore, the expression of the output feedback observer is:

[0032]

[0033] in, and is the observed value, L ri is the feedback parameter, is the r-th switching delay value of vehicle i, and the range of r is [1, M].

[0034] Furthermore, the expression of the multi-intelligent vehicle cooperative controller is:

[0035]

[0036] Among them, K ri To control the parameters, the conversion formula

[0037] Furthermore, the expression of the stabilizer is:

[0038]

[0039] in, is the observation error,

[0040] Furthermore, the stability condition of the stabilizer is:

[0041]

[0042]

[0043]

[0044]

[0045]

[0046]

[0047] in, R ri =K ri P ri , S ri =Q ri L ri , r=0,...,M.

[0048] Furthermore, the specific process of offline acquisition of several types of offline parameters of the multi-intelligent vehicle cooperative controller includes:

[0049] S431, set r=0 and Δh 0i and ε 0i The value of

[0050] S432, r, Δh 0i and ε 0i Substituting the value of into the stability equation of the multi-intelligent vehicle system, we can get P ri , Q ri 、M 1ri 、M 2ri , S ri and R ri ,calculate as well as Let r = r + 1;

[0051] S433, determine whether it is satisfied If so, Select Δh ri and ε ri , execute step S432, otherwise execute step S434;

[0052] S434, output Δh ri , ε ri , P ri , Q ri , L ri and K ri Offline parameters when r=0,...,M.

[0053] Furthermore, the specific process of the adaptive switching strategy includes:

[0054] S441, obtaining offline parameters;

[0055] S442, Order t (r+1)i is the next switching moment, when t≥p ri When , go to step S443;

[0056] S443, substituting the offline parameters into the following inequality:

[0057]

[0058] in, To adjust the constant, determine whether the inequality holds. If so, set t = t r+1 , r=r+1, execute step S444, otherwise execute step S445;

[0059] S444, determine whether r≥M is satisfied, if so, set r=0 and execute step S443, otherwise execute step S445;

[0060] Step S445: Input the current

[0061] Compared with the prior art, the present invention has the following beneficial effects:

[0062] (1) The present invention introduces unknown input delays into the kinematic model of intelligent vehicles, establishes a multi-intelligent vehicle system model based on the kinematic model of intelligent vehicles and the topological architecture of multiple intelligent vehicles, obtains observation values ​​of intelligent vehicles through output feedback observers, inputs the observation values ​​into the multi-intelligent vehicle collaborative controller, obtains several types of offline parameters of the multi-intelligent vehicle collaborative controller offline, switches several types of offline parameters online through the adaptive switching strategy of the multi-intelligent vehicle system model until the stabilizer of the multi-intelligent vehicle system model is stable, obtains the unknown input delay of each intelligent vehicle, inputs the unknown input delay into the multi-intelligent vehicle collaborative controller, and realizes longitudinal control of the multiple intelligent vehicles. When designing the multi-intelligent vehicle collaborative controller, the interference of input delay on multi-vehicle control is considered, and the specific position input delay value of each vehicle is obtained through the adaptive switching estimator, which effectively ensures the safety and stability of the longitudinal collaborative control of the intelligent vehicle fleet;

[0063] (2) The present invention only uses vehicle-mounted sensors to obtain the position information of the preceding vehicle, and obtains the speed and acceleration information of the preceding vehicle through a designed observer. Communication between vehicles is not required, thus avoiding interference caused by communication packet loss or delay, and having strong anti-interference capability. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 is a flow chart of the method of the present invention;

[0065] Figure 2 A schematic diagram of the fleet;

[0066] Figure 3 a time graph of the estimated input delay value for the following vehicle;

[0067] Figure 4 is a time graph of the position of the following vehicle;

[0068] Figure 5 is a time curve diagram of the speed of the following vehicle;

[0069] Figure 6 is a time graph of the following vehicle acceleration;

[0070] Figure 7 This is a schematic diagram of the position and speed information of the convoy at the 15th second;

[0071] Figure 8 This is a schematic diagram of the position and speed information of the fleet at the 75th second. DETAILED DESCRIPTION

[0072] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0073] An observation-based cooperative control method for multiple intelligent vehicles with unknown input delays, such as Figure 1 ,include:

[0074] S1. Based on the vehicle's dynamic equations, the intelligent vehicle is modeled taking into account the unknown input delay of the vehicle;

[0075] S2. Construct the topological architecture of multiple intelligent vehicles and establish the system model among multiple intelligent vehicles;

[0076] S3, design an output feedback observer and a vehicle controller to input the estimated observation value of the leading vehicle into the multi-intelligent vehicle cooperative controller;

[0077] S4. Design a stabilizer and adaptive switching strategy for the multi-intelligent vehicle system, obtain a series of switching data offline, switch the data online until the system is stable, and estimate the unknown input delay of each vehicle. Input the unknown input delay into the controller to achieve longitudinal control of the multi-intelligent vehicles.

[0078] In step S1, the intelligent vehicle has an unknown input delay, which can be described as:

[0079]

[0080] Among them, qi (t) represents the position of intelligent vehicle i, p i (t) represents the speed of intelligent vehicle i, a i (t) represents the acceleration of intelligent vehicle i, η represents the time constant of lag, h i represents the unknown input delay of intelligent vehicle i, and i represents the time from the 1st vehicle to the Nth vehicle.

[0081] The topological structure of the convoy in step S2 is that each vehicle only uses the on-board sensor to obtain the position information of the preceding vehicle, and the safe distance between vehicle i and preceding vehicle i-1 is set to d0, and the error between the actual distance between vehicle i and preceding vehicle i-1 and the expected distance is Δδ i (t) = q i (t)-q i-1 (t)-d0, the subsystem of vehicle i and the preceding vehicle i-1 can be described as:

[0082]

[0083] The multi-intelligent vehicle system can be described as:

[0084]

[0085] x i (t) = [Δδ i (t), Δp i (t), Δa i (t)] T

[0086] y i (t) = Δδ i (t)

[0087]

[0088]

[0089] C i =[0 0 1]

[0090] ρ i (t) = [0 0 -B i u i-1 (th i-1 )] T

[0091] The formation target in step S2 is specifically:

[0092]

[0093] Step S3 specifically includes the following steps:

[0094] S31. In order to avoid communication delay and packet loss caused by inter-vehicle communication, an observer is designed to obtain the speed and acceleration information of the preceding vehicle:

[0095]

[0096] in, and is the observed value, L ri is the feedback parameter, is the r-th switching delay value of vehicle i, the range of r is [1, M], and the observation error of the multi-intelligent vehicle system is:

[0097]

[0098] S32. Design the controller of multi-intelligent vehicle system:

[0099]

[0100] Among them, K ri To control the parameters, the conversion formula

[0101] Step S4 specifically includes the following steps:

[0102] S41. Design the stability equation of the multi-intelligent vehicle system as:

[0103]

[0104] S42. According to the stability equation of the multi-intelligent vehicle system, the stability condition of the multi-intelligent vehicle system is:

[0105]

[0106]

[0107]

[0108]

[0109]

[0110]

[0111] in, R ri =K ri P ri , S ri =Q ri L ri , the above formula has a symmetric positive definite matrix solution P ri and Q ri .

[0112] S43, calculating offline parameters, including the following sub-steps:

[0113] S431, set r=0 and Δh 0i and ε 0i The value of

[0114] S432, r, Δh 0i and ε 0i Substituting the value of into the stability equation of the multi-intelligent vehicle system, we can get P ri , Q ri 、M 1ri 、M 2ri , S ri and R ri ,calculate as well as Let r = r + 1;

[0115] S433, determine whether it is satisfied If so, Select Δh ri and ε ri , execute step S432, otherwise execute step S434;

[0116] S434, output Δh ri , ε ri , P ri , Q ri , L ri and K ri Offline parameters when r=0,...,M.

[0117] S44, setting adaptive switching strategy:

[0118] S441, obtaining a series of offline parameters obtained in sub-step S434;

[0119] S442, Order t (r+1)i is the next switching moment, when t≥p ri When , go to step S443;

[0120] S443, substituting the offline parameters into the following inequality:

[0121]

[0122] in, To adjust the constant, determine whether the inequality holds. If so, set t = t r+1 , r=r+1, execute step S444, otherwise execute step S445;

[0123] S444, determine whether r≥M is satisfied, if so, set r=0 and execute step S443, otherwise execute step S445;

[0124] Step S445: Input the current

[0125] Explain in detail with specific examples. Figure 2 As a schematic diagram of a convoy, consider a convoy of four following vehicles, with the speed of the leading vehicle p0 = 10 m / s. The convoy uses on-board sensors to obtain the position information of the leading vehicle, and the time constant of the lag is 0.5. The topology of the convoy is that each vehicle only uses on-board sensors to obtain the position information of the leading vehicle. The safety distance between vehicle i and the leading vehicle i-1 is set to d0 = 20 m. Finally, the estimated input delay of the convoy is obtained as follows: Figure 3 As shown in Figure 4 As shown, the speed of the vehicle Figure 5 As shown, as the vehicle's acceleration Figure 6 As shown, at the 15th second, the position and speed information of the fleet is as follows Figure 7 As shown, at the 75th second, the position and speed information of the fleet is as follows Figure 8 shown.

[0126] This embodiment proposes an observation-based multi-intelligent vehicle cooperative control method with unknown input delay. The relative positions between vehicles are obtained through on-board sensors, and the speed and acceleration information of the leading vehicle is obtained by an observer to avoid packet loss and delay caused by communication between vehicles. At the same time, the input delay of the vehicle is taken into account, and the position input delay of each vehicle is obtained through an adaptive switching estimator, which ultimately effectively ensures the safety and stability of the longitudinal cooperative control of the intelligent fleet.

[0127] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A cooperative control method for multiple intelligent vehicles with unknown input delays based on observation. It is characterized in that include: Construct a kinematic model of an intelligent vehicle with unknown input delays; Establish a multi-intelligent vehicle system model based on the intelligent vehicle kinematic model and the topological architecture of multiple intelligent vehicles; Obtain observation values ​​of the intelligent vehicles through the output feedback observer, and input the observation values ​​into the multi-intelligent vehicle cooperative controller; Offline acquisition of several types of offline parameters of the multi-intelligent vehicle cooperative controller, online switching of several types of offline parameters through the adaptive switching strategy of the multi-intelligent vehicle system model, until the stabilizer of the multi-intelligent vehicle system model is stable, and the unknown input delay of each intelligent vehicle is obtained; The unknown input delay is input into the multi-intelligent vehicle cooperative controller to achieve longitudinal control of the multi-intelligent vehicles; The expression of the intelligent vehicle kinematic model with unknown input delay is: Among them, q i (t) represents the position of intelligent vehicle i, p i (t) represents the speed of intelligent vehicle i, a i (t) represents the acceleration of intelligent vehicle i, η represents the time constant of lag, h i represents the unknown input delay of intelligent vehicle i, where i represents the time from the first vehicle to the Nth vehicle; In the topological architecture of multiple intelligent vehicles, each vehicle only uses the on-board sensor to obtain the position information of the preceding vehicle. The subsystem expressions of the vehicle i and the preceding vehicle i-1 are: Among them, Δδ i (t) is the error between the actual distance and the expected distance between vehicle i and the preceding vehicle i-1, Δδ i (t) = q i (t)-q i-1 (t)-d 0 , d 0 is the safe distance between vehicle i and the preceding vehicle i-1; The expression of the multi-intelligent vehicle system model is: x i (t)=[Δδ i (t),Δp i (t), Δa i (t)] T y i (t)=Δδ i (t) ρ i (t)=[00-B i u i-1 (t-h i-1 )] T ; The expression of the output feedback observer is: in, and is the observed value, L ri is the feedback parameter, is the r-th switching delay value of vehicle i, and the range of r is [1, M]; The expression of the multi-intelligent vehicle cooperative controller is: Among them, K ri To control the parameters, the conversion formula 2. According to the observation-based multi-intelligent vehicle cooperative control method with unknown input delay according to claim 1, the formation goal of the multi-intelligent vehicle system model is specifically:

3. According to the observation-based multi-intelligent vehicle cooperative control method with unknown input delay in claim 1, the expression of the stabilizer is: in, is the observation error, 4. According to the observation-based multi-intelligent vehicle cooperative control method with unknown input delay according to claim 3, the stability condition of the stabilizer is: in, R ri =K ri P ri ,S ri =Q ri L ri ,r=0,...,M。 5. According to the observation-based multi-intelligent vehicle cooperative control method with unknown input delay according to claim 4, the specific process of offline acquisition of several types of offline parameters of the multi-intelligent vehicle cooperative controller include: S431, set r=0 and Δh 0i and ε 0i The value of S432, r, Δh 0i and ε 0i Substituting the value of into the stability equation of the multi-intelligent vehicle system, we can get P ri , Q ri 、M 1ri 、M 2ri , S ri and R ri ,calculate as well as Let r = r + 1; S433, determine whether it is satisfied If so, Select Δh ri and ε ri , execute step S432, otherwise execute step S434; S434, output Δh ri , ε ri , P ri , Q ri , L ri and K ri Offline parameters when r=0,...,M.

6. According to the observation-based multi-intelligent vehicle cooperative control method with unknown input delay according to claim 5, the specific process of the adaptive switching strategy include: S441, obtaining offline parameters; S442, Order t (r+1)i is the next switching moment, when t≥p ri When , go to step S443; S443, substituting the offline parameters into the following inequality: in, To adjust the constant, determine whether the inequality holds. If so, set t = t r+1 , r=r+1, execute step S444, otherwise execute step S445; S444, determine whether r≥M is satisfied, if so, set r=0 and execute step S443, otherwise execute step S445; Step S445: Input the current