An environmental detection method based on autonomous dynamic formation

Abstract

The present invention discloses an environment detection method based on autonomous dynamic formation, comprising the following steps: S1, determining a multi-agent system set, denoting the total number of intelligent vehicles as n, and establishing a communication network topology graph G of the multi-agent system; S2, constructing an intelligent vehicle system model in a moving local coordinate system; S3, designing a formation control protocol that does not depend on the acceleration of the leader vehicle for the case of the leader vehicle's acceleration; S4, writing the designed formation control protocol into each intelligent vehicle through programming, and realizing distributed information interaction among the intelligent vehicles through the established communication topology graph G, constructing a closed-loop system in a compact form, and realizing the given formation control; S5, driving the multi-agent formation by the leader vehicle through controlling its own acceleration to perform autonomous coverage and complete the coverage and detection operations of the task environment. This method improves the detection accuracy, reduces the requirement for the equipment and data processing capabilities of each intelligent vehicle, and thus reduces the cost.

Description

Technical Field

[0001] The present invention relates to the field of computer science and control technology, and particularly to an environment detection method based on autonomous dynamic formation. Background Art

[0002] Environmental monitoring refers to the activities of environmental monitoring agencies to monitor and measure the environmental quality status. By monitoring and measuring the indicators reflecting the environmental quality, it is determined the pollution status of the environment and the level of environmental quality. The content of environmental monitoring includes both the detection of chemical pollutants and the monitoring of physical (energy) factors such as noise, vibration, thermal energy, electromagnetic radiation, and radioactivity; and also includes biological monitoring of various reactions and information emitted by organisms due to changes in environmental quality, as well as ecological monitoring of the migration changes of regional communities and populations, etc.

[0003] In recent years, the problem of coverage and detection of the mission environment has gradually attracted the attention of the academic community. The traditional design method designs a wireless sensor network, calculates the optimal coverage nodes, and places the sensors manually; however, this method has disadvantages such as inflexibility and poor robustness, so it has gradually been replaced by the multi-agent detection method. This method uses multi-agents to carry corresponding sensors, autonomously configure and move to the optimal coverage nodes to complete environmental detection.

[0004] However, this detection method also has certain defects: Since each agent is a relatively independent individual, it depends on the communication radius during the coverage process, and for each agent relative to the group, the sensing radius of the sensors carried by it is small and the detection value is inaccurate. Therefore, forming a formation by multiple agents, expanding their communication radius and sensing radius, and improving their detection accuracy through data fusion, and performing autonomous coverage, is a feasible method.

[0005] However, the resulting formation control problem is one of the most important problems in multi-agent systems. Early formation control was mostly traditional distributed consensus algorithms, which relied on some global information, such as whether the base axes of the local coordinate systems of the agents were aligned, which was difficult to guarantee in actual engineering. In view of the above defects, scholars have proposed two different formation control algorithms: the formation control algorithm based on rigid graph theory and the circular formation control algorithm. However, both methods have their own defects. Among them, it is difficult to extend the formation control method based on rigid graph theory to general linear systems. And in the research method of circular formation control, the existing algorithms rely on each agent to accurately and real-time measure the acceleration of the target and the leader, which increases the requirements for the equipment and data processing capabilities of the agents, and then requires the agents to be equipped with expensive and high-precision sensors and processing units. Summary of the Invention

[0006] In view of the deficiencies of the prior art, the present invention proposes an environment detection method based on autonomous dynamic formation, which combines a coverage control algorithm and a formation control algorithm, expands the communication radius and sensing radius, and improves the detection accuracy through data fusion.

[0007] To solve the above technical problems, the technical solution of the present invention is as follows:

[0008] An environment detection method based on autonomous dynamic formation, comprising the following steps:

[0009] S1. Determine the multi-agent system set, denote the total number of intelligent vehicles as n, establish the communication network topology graph G of the multi-agent system, and the i-th intelligent vehicle has exactly two neighbors i + and i - ;

[0010] S2. Construct the intelligent vehicle system model in the moving local coordinate system:

[0011] S3. For the case of the acceleration of the leader vehicle, design a formation control protocol that does not depend on the acceleration of the leader vehicle;

[0012] S4. Write the designed formation control protocol into each intelligent vehicle through programming, and realize the distributed information interaction between intelligent vehicles through the established communication topology graph G, construct a closed-loop system in a compact form, and realize the given formation control;

[0013] S5. The leader vehicle drives the multi-agent formation by controlling its own acceleration to perform autonomous coverage and complete the coverage and detection operations of the task environment.

[0014] Preferably, in the multi-agent system set, the i-th intelligent vehicle has exactly two neighbors i + and i - .

[0015] Preferably, in step S2, the construction of the intelligent vehicle system model is specifically as follows:

[0016]

[0017] where ρ i and η i respectively represent the relative distance and relative distance speed between the i-th intelligent vehicle and the leader vehicle, represents the relative angle between the i-th intelligent vehicle and its neighbor i + and ω i , ω i+ and ω i- respectively represent the i-th intelligent vehicle and its neighbors i + , i -Around the angular velocity of the leader vehicle respectively, and respectively represent the accelerations of the leader vehicle acting on the x-axis and y-axis of the moving local coordinate system, and respectively represent the controller inputs acting on the x-axis and y-axis of the moving local coordinate system.

[0018] Preferably, the intelligent vehicle system model is transformed from a second-order integral multi-intelligent vehicle system in the global coordinate system and a second-order integral dynamic system of the leader vehicle,

[0019] wherein, the second-order integral multi-intelligent vehicle system in the global coordinate system is as follows:

[0020]

[0021] The second-order integral dynamic system of the leader vehicle is as follows:

[0022]

[0023] where p i (t), v i (t) and u i (t) respectively represent the position, velocity and controller input of the i-th intelligent vehicle at time t in the global coordinate system, p0(t), v0(t) and u0(t) respectively represent the position, velocity and acceleration of the leader vehicle at time t in the global coordinate system, where a0(t) cannot be directly measured but has an upper bound

[0024] Preferably, in step S3, for the known upper bound of the leader vehicle acceleration a ring formation control protocol that does not depend on the leader vehicle acceleration a0(t) is designed:

[0025]

[0026] The control variables need to satisfy:

[0027]

[0028] where, R i represents the desired relative distance between the i-th intelligent vehicle and the leader vehicle, d i represents the desired relative angle between the i-th intelligent vehicle and its neighbor i + Ω represents the desired angular velocity of each intelligent vehicle around the leader vehicle,

[0029] The control parameters l1 and l2 are selected to satisfy:

[0030]

[0031] Preferably, in the step S3, for the upper bound of the acceleration of the leader vehicle unknown situation, design an adaptive circular formation control protocol that does not depend on the acceleration a0(t) of the leader vehicle:

[0032]

[0033] The control variables need to satisfy:

[0034]

[0035] where, R i represents the desired relative distance between the i-th intelligent vehicle and the leader vehicle, d i represents the desired relative angle between the i-th intelligent vehicle and its neighbor i + and Ω represents the desired angular velocity of each intelligent vehicle around the leader vehicle.

[0036] And design the adaptive coefficient update law:

[0037]

[0038] where, e i1 and e i2 represent the adaptive coefficients, the constants τ i1 > 0 and τ i2 > 0.

[0039] Preferably, in the step S4, for the case where the upper bound of the acceleration is known, the constructed compact-form closed-loop system is:

[0040]

[0041] where, ρ = [ρ1, ρ1,..., ρ n T , ω = [ω1, ω1,..., ω n T , η = [η1, η1,..., η n T ,

[0042] ω + = [ω2, ω3,..., ω n , ω1] T , f = [f1, f1,..., f n T ​​​​Vectors representing each variable respectively; diag(·) represents diagonalizing a vector, and sign(·) represents performing a sign function operation on each element in the vector.

[0043]

[0044] Preferably, in the step S4, for the upper bound of acceleration In the case of unknown, the constructed compact - form closed - loop system is:

[0045]

[0046] where e1 = [e 11 , e 21 ,..., e n1 T , e2 = [e 12 , e 22 ,..., e n2 T represent the vector forms of the adaptive coefficients.

[0047] Preferably, in the step S4, the method for achieving the given circular formation control:

[0048] Leader's circum - navigation:

[0049]

[0050] Distributed regulation:

[0051]

[0052] where R = [R1, R1,..., R n T and d = [d1, d1,..., d n T represent the vector forms of each variable respectively; the symbol · * represents the state that each state is expected to reach.

[0053] The present invention has the following characteristics and beneficial effects:

[0054] An environmental monitoring method combining formation control and coverage control is proposed, which solves the problems of insufficient communication radius, sensing radius, and detection accuracy; two circular formation control protocols independent of the target's real - time acceleration are proposed, thus reducing the requirements for the equipped devices and data - processing capabilities of each intelligent vehicle. Therefore, the intelligent vehicle does not need to be equipped with expensive and high - precision sensors and processing units, thereby reducing costs.

[0055] ​​​​Based on the distributed idea that the intelligent vehicle only relies on local information, according to the conversion relationship between the global coordinate system and the moving local coordinate system, the dynamic system of each intelligent vehicle described in the local coordinate system is constructed, and the local state information obtained is used for the designed circular formation control protocol, thus avoiding the use of global information by the intelligent vehicle and improving the robustness of the multi-agent system.

[0056] Furthermore, the second formation control protocol is improved on the basis of the first formation control protocol, so that the intelligent vehicle no longer depends on the upper bound information of the target acceleration, thus enabling the intelligent vehicle to avoid manually resetting the control parameters according to the upper bound each time and expanding the applicable range. Brief Description of the Drawings

[0057] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0058] Figure 1 It is the principle flowchart of the embodiment of the present invention.

[0059] Figure 2 It is a schematic diagram of the intelligent vehicle surrounding the leader vehicle in the embodiment of the present invention;

[0060] Figure 3 It is a schematic diagram of the formation trajectory in the embodiment of the present invention. Detailed Embodiments

[0061] It should be noted that, without conflict, the embodiments in the present invention and the features in the embodiments can be combined with each other.

[0062] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "transverse", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus cannot be understood as a limitation to the present invention. In addition, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first", "second", etc. may explicitly or implicitly include one or more of such features. In the description of the present invention, unless otherwise stated, the meaning of "a plurality" is two or more.

[0063] In the description of the present invention, it should be noted that unless otherwise clearly specified and defined, the terms "installation", "connection", and "coupling" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, and it can be the communication inside two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0064] Embodiment 1

[0065] This embodiment provides an environment detection method based on autonomous dynamic formation, as Figure 1 shown, including the following steps:

[0066] S1. Determine the multi-agent system set. Denote the total number of intelligent vehicles as n, and establish a communication network topology graph G of the multi-agent system. The i-th intelligent vehicle has exactly two neighbors i + and i - .

[0067] Specifically, denote the total number of intelligent vehicles as n. The communication topology of the intelligent vehicle group can be described as an undirected connected graph G(V, E), where V = {v1,... v n} and E represent the point set and the edge set respectively. For an undirected graph, the edge ∈ ij ∈ E represents that the i-th intelligent vehicle and the j-th intelligent vehicle can transmit information to each other. N i = {j|∈ ij ∈ E, j ≠ i} represents the communication neighbor set of the i-th intelligent vehicle.

[0068] In particular, for each intelligent vehicle i, there are exactly two neighbors N i = {i + , i -}, where

[0069]

[0070] and

[0071]

[0072] S2. Construct an intelligent vehicle system model in the moving local coordinate system.

[0073] Specifically, as Figure 2 shown, construct an intelligent vehicle system model as follows:

[0074] Let a i(t) ∈ [0, 2π] represents the angle of the i-th intelligent vehicle in the global coordinate system, and we have

[0075] 0 ≤ α1(0) < α2(0) < … α n (0) < 2π

[0076] Let represent the relative angle between the i-th intelligent vehicle and its neighbor i + Then we have

[0077]

[0078] Therefore, we have And due to the actual situation is continuous, then holds at any time t.

[0079] Given that the second-order integral multi-intelligent vehicle system can be transformed into an intelligent vehicle system model in the moving local coordinate system:

[0080]

[0081] where ρ i and η i respectively represent the relative distance and relative distance velocity between the i-th intelligent vehicle and the leader vehicle, represents the relative angle between the i-th intelligent vehicle and its neighbor i + ω i , ω i+ and ω i- respectively represent the angular velocities of the i-th intelligent vehicle and its neighbors i + , i - respectively around the leader vehicle, and respectively represent the accelerations of the leader vehicle acting on the x-axis and y-axis of the moving local coordinate system, and respectively represent the controller inputs acting on the x-axis and y-axis of the moving local coordinate system.

[0082] It should be noted that the local relative state information ρ i , η i , ω i , ω i+ and ω i- are obtained through measurement, calculation, and communication with neighbors in the moving local coordinate system.

[0083] Furthermore, the intelligent vehicle system model is transformed from the second-order integral multi-intelligent vehicle system in the global coordinate system and the second-order integral dynamic system of the leader vehicle,

[0084] Among them, the second-order integral multi-intelligent vehicle system in the global coordinate system is as follows:

[0085]

[0086] The second-order integral dynamic system of the leader vehicle is as follows:

[0087]

[0088] Where p i (t), v i (t) and u i (t) represent the position, velocity and controller input of the i-th intelligent vehicle at time t in the global coordinate system respectively, p0(t), v0(t) and u0(t) represent the position, velocity and acceleration of the leader vehicle at time t in the global coordinate system respectively, where a0(t) cannot be directly measured but has an upper bound

[0089] Where and represent the position, position and controller input of the i-th intelligent vehicle at time t in the global coordinate system respectively; and represent the position, position and acceleration of the i-th intelligent vehicle at time t in the global coordinate system respectively; In particular, the and calculated from a0 also have the properties of being not directly measurable, but Lebesgue measurable and bounded, and satisfy the following inequalities:

[0090]

[0091] The specific conversion process is as follows:

[0092] Let and represent the relative position and relative velocity between the i-th intelligent vehicle and the leader vehicle in the global coordinate system respectively; Further, we can obtain the following dynamic system:

[0093]

[0094] For each intelligent vehicle i, a moving local coordinate system is constructed. In particular, the x-axis points from the position p0(t) of the leader vehicle to the position p i (t) of the i-th intelligent vehicle. Therefore, we can obtain the mutual relationship between the global coordinate system and the moving local coordinate system:

[0095]

[0096] Where and respectively represent the local relative position and local relative velocity between the \(i\)-th intelligent vehicle and the leader vehicle in the moving local coordinate system, and \(A\) i (t) represents the rotation matrix and has

[0097]

[0098] In particular, there is

[0099] According to the above mutual relationships, we can obtain:

[0100]

[0101] where represents the acceleration of the leader vehicle in the moving local coordinate system of the \(i\)-th intelligent vehicle, and there is

[0102] Finally, according to the above equations, the intelligent vehicle system model in the given moving local coordinate system of this embodiment is obtained.

[0103] S3. For the case of the acceleration of the leader vehicle, design a formation control protocol that does not depend on the acceleration of the leader vehicle,

[0104] Specifically, for the case where the upper bound of the acceleration of the leader vehicle is known, design a circular formation control protocol that does not depend on the acceleration \(a_0(t)\) of the leader vehicle:

[0105]

[0106] The control variables need to satisfy:

[0107]

[0108] where \(R\) i represents the expected relative distance between the \(i\)-th intelligent vehicle and the leader vehicle, \(d\) i represents the expected relative angle between the \(i\)-th intelligent vehicle and its neighbor \(i\) + and \(\Omega\) represents the expected angular velocity of each intelligent vehicle around the leader vehicle, \(R\) i > 0, \(d\) i > 0,

[0109] The control parameters \(l_1\) and \(l_2\) selected need to satisfy:

[0110]

[0111] S4. Write the designed formation control protocol into each intelligent vehicle through programming, and realize the distributed information interaction among intelligent vehicles through the established communication topology graph G, construct a closed-loop system in a compact form, and realize the given formation control;

[0112] Specifically, in the step S4, for the known upper bound of acceleration , the constructed closed-loop system in a compact form is:

[0113]

[0114] where ρ = [ρ1, ρ1,..., ρ n T , ω = [ω1, ω1,..., ω n T , η = [η1, η1,..., η n T ,

[0115] ω + = [ω2, ω3,..., ω n , ω1] T , f = [f1, f1,..., f n T respectively represent the vector forms of each variable; diag(·) represents diagonalizing the vector, and sign(·) represents performing the sign function operation on each element in the vector.

[0116]

[0117] Furthermore, in the step S4, the method for realizing the given circular formation control:

[0118] Leader's circumvolution:

[0119]

[0120] Distributed regulation:

[0121]

[0122] where R = [R1, R1,..., R n T and d = [d1, d1,..., d n T respectively represent the vector forms of each variable; the symbol · * represents the state that each state is expected to reach.

[0123] ​​​​​​It is understandable that for the constructed compact - form closed - loop system, as long as the control - parameter conditions are met, there is

[0124] 1) The system state converges to the sliding - mode surface \(s(t)=0\) within a specified time, and the upper bound of this time is \(l = \min\{l_1,l_2\}\) and

[0125]

[0126] For \(y(t)=[y_1(t),y_2(t),\cdots,y n (t)] T , there is \(y i (t)=[\omega i (t)-\Omega - f i (t)]\rho i (t);

[0127] 2) The state of the compact - form closed - loop system converges to the desired formation at an exponential rate

[0128] S5. The leader vehicle drives the multi - agent formation by controlling its own acceleration to perform autonomous coverage and complete the coverage and detection operations of the task environment.

[0129] Embodiment 2

[0130] The difference between this embodiment and the previous embodiment is that in step S3, for the case where the upper bound of the acceleration of the leader vehicle is unknown, an adaptive circular - formation control protocol that does not depend on the acceleration \(a_0(t)\) of the leader vehicle is designed:

[0131]

[0132] The control variables need to satisfy:

[0133]

[0134] where \(R i represents the desired relative distance between the \(i\) - th intelligent vehicle and the leader vehicle, \(d i represents the desired relative angle between the \(i\) - th intelligent vehicle and its neighbor \(i + ), \(\Omega\) represents the desired angular velocity of each intelligent vehicle around the leader vehicle,

[0135] And an adaptive - coefficient update law is designed:

[0136]

[0137] where \(e i1 and \(e i2 represent the adaptive coefficients, and the constant \(\taui1 > 0 and τ i2 > 0.

[0138] Furthermore, in step S4, for the case where the upper bound of acceleration is unknown, the constructed compact - form closed - loop system is:

[0139]

[0140] where e1 = [e 11 , e 21 ,..., e n1 T , e2 = [e 12 , e 22 ,..., e n2 T represents the vector form of the adaptive coefficient.

[0141] 3) It can be understood that, as Figure 3 shown, for the compact - form closed - loop system constructed in Embodiment 2, the state of the compact - form closed - loop system asymptotically converges to the desired formation

[0142] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. For those skilled in the art, without departing from the principle and spirit of the present invention, various changes, modifications, substitutions, and variations to these embodiments including components still fall within the protection scope of the present invention.​​

Claims

1. An environmental detection method based on autonomous dynamic formation, characterized in that It includes the following steps: S1. Determine the multi-agent system set, denote the total number of intelligent vehicles as n, and establish the communication network topology graph G of the multi-agent system; S2. Construct the intelligent vehicle system model in the moving local coordinate system: Construct the intelligent vehicle system model as follows: Among them, ρ i and η i respectively represent the relative distance and the relative distance speed between the i-th intelligent vehicle and the leader vehicle, represents the relative angle between the i-th intelligent vehicle and its neighbor i + ω i and ω i+ ω i- respectively represent the angular velocities of the i-th intelligent vehicle and its neighbor i + around the leader vehicle respectively, and respectively represent the accelerations of the leader vehicle acting on the x-axis and y-axis of the moving local coordinate system, and respectively represent the controller inputs acting on the x-axis and y-axis of the moving local coordinate system; S3. For the case of the acceleration of the leader vehicle, design a formation control protocol that does not depend on the acceleration of the leader vehicle; For the upper bound of the acceleration of the leader vehicle For the known situation, a circular formation control protocol that does not depend on the acceleration a0(t) of the leader vehicle is designed: The control variables need to satisfy: where R i represents the desired relative distance between the i-th intelligent vehicle and the leader vehicle, d i represents the desired relative angle between the i-th intelligent vehicle and its neighbor i + and Ω represents the desired angular velocity of each intelligent vehicle around the leader vehicle. The selected control parameters l1 and l2 need to satisfy: For the case where the upper bound of the acceleration of the leader vehicle is unknown, an adaptive circular formation control protocol that does not depend on the acceleration a0(t) of the leader vehicle is designed: The control variables need to satisfy: where, R i represents the expected relative distance between the i-th intelligent vehicle and the leader vehicle, d i represents the expected relative angle between the i-th intelligent vehicle and its neighbor i + and Ω represents the expected angular velocity of each intelligent vehicle around the leader vehicle And design the adaptive coefficient update law: Among them, e i1 and e i2 represent the adaptive coefficients, the constant τ i1 > 0 and τ i2 > 0; S4. Write the designed formation control protocol into each intelligent vehicle through programming, and realize the distributed information interaction among the intelligent vehicles through the established communication topology graph G, construct a closed-loop system in a compact form, and realize the given formation control; S5. The leader vehicle drives the multi-agent formation by controlling its own acceleration to perform autonomous coverage and complete the coverage and detection operations of the task environment.

2. The environmental detection method based on autonomous dynamic formation according to claim 1, characterized in that In the set of the multi-agent systems, the $i$-th intelligent vehicle has exactly two neighbors $i$ + and $i$ - .

3. The environmental detection method based on autonomous dynamic formation according to claim 2, wherein The intelligent vehicle system model is transformed from the second-order integral multi-intelligent vehicle system in the global coordinate system and the second-order integral dynamic system of the leader vehicle, wherein, the second-order integral multi-intelligent vehicle system in the global coordinate system is as follows: The second-order integral dynamic system of the leader vehicle is as follows: where p i (t), v i (t) and u i (t) respectively represent the position, velocity and controller input of the i-th intelligent vehicle at time t in the global coordinate system. p0(t), v0(t) and u0(t) respectively represent the position, velocity and acceleration of the leader vehicle at time t in the global coordinate system, where a0(t) cannot be directly measured but has an upper bound 4. The environmental detection method based on autonomous dynamic formation according to claim 3, characterized in that In the step S4, for the known upper bound of acceleration the constructed compact - form closed - loop system is: where ρ = [ρ1, ρ1,..., ρ n T , ω = [ω1, ω1,..., ω n T , η = [η1, η1,..., η n T ,​​​ ω + = [ω2, ω3,..., ω n , ω1] T , f = [f1, f1,..., f n T ​ respectively represent the vector forms of each variable; diag(·) represents diagonalizing the vector, and sign(·) represents performing the sign function operation on each element in the vector, 。 5. The environmental detection method based on autonomous dynamic formation according to claim 4, characterized in that In the step S4, for the upper bound of acceleration when it is unknown, the constructed compact form closed-loop system is: Among them, e1 = [e 11 , e 21 ,..., e n1 T , e2 = [e 12 , e 22 ,..., e n2 T represents the vector form of the adaptive coefficient.​​ 6. The environmental detection method based on autonomous dynamic formation according to claim 4 or 5, characterized in that, In step S4, the method for realizing the given circular formation control: Leader circumvolution: Distributed regulation: where R = [R1, R1,..., R n T and d = [d1, d1,..., d n T represent the vector forms of the respective variables; the symbol · * represents the state that each state is expected to reach.​​

Images