A Multi-Agent Indoor Cooperative Localization Method Based on Artificial Waymarks and Sound Technology

Abstract

This invention discloses a multi-agent indoor collaborative localization method based on artificial landmarks and acoustic technology, belonging to the field of indoor positioning technology. This invention proposes a multi-agent collaborative localization method based on artificial landmarks and acoustic technology. By combining acoustic technology and machine vision, it fuses angle, distance, and relative motion speed information among multiple agents to achieve collaborative localization. This method has high position estimation accuracy, low application cost, and high system compatibility. Compared with traditional methods, this invention achieves collaborative localization by fusing acoustic technology. Since acoustic technology only requires microphones and speakers, it has lower application costs and higher system compatibility, thus possessing greater practical application value.

Description

Technical Field

[0001] This invention belongs to the field of indoor positioning technology, specifically relating to a multi-agent indoor collaborative positioning method based on artificial landmarks and sound technology. Background Technology

[0002] Intelligent systems have been widely applied in various fields such as production, daily life, healthcare, and the military. For example, intelligent hospitals can use intelligent robots to provide services such as meal delivery and medication dispensing for special types of patients, especially during the COVID-19 pandemic, when contactless delivery has become particularly important. Therefore, research on positioning methods, as a fundamental engineering aspect of intelligent systems, has become a hot topic.

[0003] Since intelligent agents are generally equipped with cameras, visual positioning has become one of the mainstream methods for indoor positioning at present. In complex indoor environments, intelligent agents can obtain their own pose information by using landmark information collected by cameras. When using natural landmarks for positioning, the positioning results may be greatly affected by environmental changes, such as light intensity and obstructions. In contrast, artificial landmarks are designed and placed by humans, and environmental factors can be considered during the design process. The shape and encoding of artificial landmarks can be designed specifically to improve the recognition speed and accuracy of intelligent agents. However, single intelligent agents often find it difficult to accurately and efficiently determine their own position when facing complex and changing environments. Robots in multi-agent systems can use other agents to correct and optimize their own position, thereby achieving more accurate positioning. Moreover, the cooperation between multiple intelligent agents has stronger adaptability and robustness. Therefore, research on multi-agent cooperative positioning methods is of positive significance for the accurate positioning of intelligent systems and the future development of intelligent processes. When a single intelligent agent moves, it is limited by the number of sensors it carries, resulting in deviations in the recognition of external features, making it almost impossible to achieve high-precision positioning. The emergence of multi-agent systems allows for the fusion of information from multiple agents, achieving a "1+1>2" positioning effect. Therefore, research in multi-agent systems is receiving increasing attention from scholars both domestically and internationally. However, current methods largely rely on navigation technologies based on artificial landmarks. The fusion of information from these technologies with inertial navigation, laser technology, and UWB (Ultra-Wideband) is crucial for achieving collaborative positioning among multiple agents. Furthermore, the measurement of relative motion states between agents is primarily based on laser or UWB technology, which often results in high costs, poor compatibility, and low positioning accuracy in real-world scenarios. Summary of the Invention

[0004] In order to overcome the shortcomings of the prior art, the present invention aims to provide a multi-agent indoor collaborative positioning method based on artificial landmarks and sound technology, so as to solve the technical problems of high cost, poor compatibility and low positioning accuracy in real-world scenarios when using laser or UWB technology to achieve multi-agent collaborative positioning.

[0005] To achieve the above objectives, the present invention employs the following technical solution:

[0006] A multi-agent indoor collaborative localization method based on artificial landmarks and sound technology includes:

[0007] Set up manual road signs;

[0008] Acquire images of human-made road signs and intelligent agents;

[0009] The acquired images of artificial road signs and intelligent agents are analyzed and processed to obtain angle information;

[0010] Acquire information on the distance and relative speed between intelligent agents;

[0011] By integrating the distance, relative speed, and angle information between the agents, collaborative localization among multiple agents can be achieved.

[0012] Preferably, the setting of artificial road signs creates a position triangle between the artificial road signs and the intelligent agent.

[0013] Preferably, the images of the artificial road signs and the intelligent agent are acquired by equipping the intelligent agent with a camera.

[0014] Preferably, the angle information includes the pose angle information between the intelligent agent and the artificial road sign, as well as the angle information between the intelligent agents.

[0015] Preferably, the angle includes: the offset angle of the artificial road sign within the field of view of the intelligent agent, the offset angle of the intelligent agent within the field of view of another intelligent agent, and the distortion angle of the artificial road sign image.

[0016] Preferably, the analysis and processing of the acquired artificial road sign and agent images to obtain angle information specifically includes:

[0017] S1: Camera calibration and input image, determining the basic parameters of the input image;

[0018] S2: Perform image enhancement and binarization based on the basic parameters of the input image;

[0019] S3: Extraction of features from artificial road sign images;

[0020] S4: Recognition and extraction of intelligent agent images, autonomously constructing and labeling intelligent agent datasets, and obtaining the centroid position of intelligent agents;

[0021] S5: Based on the features of the artificial road sign image and the centroid position of the agent, the angle is solved to obtain the offset angle and distortion angle.

[0022] Preferably, the distance and relative speed information between the intelligent agents are measured using acoustic technology.

[0023] Preferably, the sound information used to measure the distance and relative speed between intelligent agents is acquired through a sound sensor and a voice module, specifically including:

[0024] S6: The sound source base station periodically broadcasts sound signals into the environment, and the receiver collects the sound signals;

[0025] S7: Perform generalized cross-correlation on the acoustic signal, process the two hyperbolic frequency modulated signal components in the signal, and obtain the arrival time by using the fixed threshold method;

[0026] S8: Perform HFM signal composite on the arrival time to obtain the distance between agents and relative velocity.

[0027] Preferably, the fusion of the acquired distance and relative motion speed information and angle information between intelligent agents to achieve cooperative localization among multiple intelligent agents specifically includes:

[0028] At time k, the agent's motion state is X. k ={x a,k ,v a,k x b,k ,v b,k} T The prediction equation and state update equation of the agent are as follows:

[0029]

[0030] Among them, F k Here is the state transition matrix.

[0031]

[0032] Y k For observations

[0033] Y k =[d k VR k ,v k ,θ 1,k ,θ 2,k ,β 1,k ,β 2,k ,α 1,k ,α 2,k ] T

[0034] h(X k ) is the observed value function; w k For predicting process noise; n k The noise in the observation process; the position x of agent a. a,k =(x a,k ,y a,k ) T The speed of agent a is v a,k =(v a,x,k ,v a,y,k ) T The position x of agent b b,k =(x b,k ,y b,k ) T The speed of agent b is v b,k =(v b,x,k ,v b,y,k ) T The sampling interval is Δt.

[0035] This invention also discloses a multi-agent indoor collaborative positioning system based on artificial landmark and sound technology, comprising:

[0036] A setting unit is used to set up manual road signs;

[0037] Image acquisition unit, used to acquire images of artificial road signs and intelligent agents;

[0038] The analysis unit is used to analyze and process the acquired images of artificial road signs and intelligent agents to obtain angle information;

[0039] The distance and velocity acquisition unit is used to acquire distance and relative motion velocity information between intelligent agents;

[0040] The fusion unit is used to fuse the distance and relative motion speed information and angle information between the acquired agents to achieve cooperative localization among multiple agents.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] This invention proposes a multi-agent cooperative localization method based on artificial landmarks and acoustic technology. By combining acoustic technology and machine vision, it integrates angle, distance, and relative motion velocity information among multiple agents to achieve cooperative localization. This method boasts high position estimation accuracy, low application cost, and high system compatibility. Compared to traditional methods, this invention achieves cooperative localization through the fusion of acoustic technology. Since acoustic technology only requires microphones and speakers, it offers lower application costs and higher system compatibility, resulting in greater practical application value. Attached Figure Description

[0043] Figure 1This is a schematic diagram illustrating the collaborative positioning principle based on artificial landmarks of the present invention;

[0044] Figure 2 This is a schematic diagram of multi-agent cooperative localization according to the present invention;

[0045] Figure 3 This is a flowchart of the angle estimation process of the present invention;

[0046] Figure 4 This is a schematic diagram of sound information acquisition according to the present invention;

[0047] Figure 5 This is a flowchart illustrating the estimation process for distance and relative velocity in this invention.

[0048] Figure 6 This is a simulation result diagram of the dynamic multi-agent cooperative localization agent a according to an embodiment of the present invention;

[0049] Figure 7 The simulation results of the dynamic multi-agent cooperative localization agent b in this embodiment of the invention are shown in the figure.

[0050] Figure 8 This is the CDF diagram of agent a in the dynamic multi-agent cooperative localization embodiment of the present invention;

[0051] Figure 9 This is the CDF diagram of agent b in the dynamic multi-agent cooperative localization embodiment of the present invention.

[0052] Wherein, 1-first intelligent agent; 2-second intelligent agent; 3-human road sign Detailed Implementation

[0053] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0054] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0055] The present invention will now be described in further detail with reference to the accompanying drawings:

[0056] Therefore, this invention addresses the problem of multi-agent localization and navigation in complex indoor environments by proposing a multi-agent cooperative localization method based on artificial landmarks and acoustic technology. First, by analyzing and processing landmark and agent images acquired by cameras mounted on the agents, pose angle information is obtained. Then, distance and relative velocity information between agents are obtained through acoustic signals. Finally, multi-source information is fused using particle filtering to achieve cooperative localization among the agents. This invention improves the localization accuracy of multi-agent systems, and also has higher compatibility and versatility, as well as lower application costs.

[0057] This invention discloses a multi-agent indoor collaborative localization method based on artificial landmarks and sound technology, comprising:

[0058] Set up manual road signs;

[0059] Acquire images of human-made road signs and intelligent agents;

[0060] The acquired images of artificial road signs and intelligent agents are analyzed and processed to obtain angle information;

[0061] Acquire information on the distance and relative speed between intelligent agents;

[0062] By integrating the distance, relative speed, and angle information between the agents, collaborative localization among multiple agents can be achieved.

[0063] This invention obtains distance and relative motion velocity measurements between multiple intelligent agents based on sound, and integrates pose information obtained based on artificial landmarks to achieve joint estimation of the positions of multiple intelligent agents. It has higher compatibility and versatility, as well as lower application costs.

[0064] In some embodiments, artificial landmarks are set up to create a position triangle between the artificial landmarks and the intelligent agent.

[0065] Figure 1 This is a schematic diagram illustrating the positioning principle of the present invention. Figure 1 As shown, a position triangle is constructed using artificial landmarks, a first intelligent agent, and a second intelligent agent. Angle information is measured via a vision system, including the pose angles between the intelligent agent and the artificial landmarks, as well as the angles between the intelligent agents themselves. Distance and relative velocity information are measured using acoustic technology. Furthermore, collaborative localization among multiple intelligent agents is achieved by fusing distance, relative velocity, and angle information.

[0066] In some embodiments, the images of the artificial road signs and the intelligent agent are acquired by equipping the intelligent agent with a camera.

[0067] like Figure 2 As shown, this invention uses a vision system, i.e., an intelligent agent equipped with a camera, to collect image information in order to obtain the angular relationship between the first intelligent agent, the second intelligent agent, and the artificial road sign, thereby constructing a position triangle.

[0068] In some embodiments, the angle information includes the pose angle information between the intelligent agent and the artificial road sign, as well as the angle information between the intelligent agents.

[0069] The angles mainly include: the offset angle θ of the artificial road sign within the agent's field of view, the offset angle β of the agent within another agent's field of view, and the distortion angle α of the artificial road sign image. Here, θ and β are the angles of offset between the centers of the artificial road sign and the agent in the image relative to the image center, and α is the distortion angle of the artificial road sign caused by the near-large-far-small phenomenon during the imaging process.

[0070] In some embodiments, the analysis and processing of the acquired artificial landmark and agent images to obtain angle information specifically includes:

[0071] S1: Camera calibration and input image, determining the basic parameters of the input image;

[0072] S2: Perform image enhancement and binarization based on the basic parameters of the input image;

[0073] S3: Extraction of features from artificial road sign images;

[0074] S4: Recognition and extraction of intelligent agent images, autonomously constructing and labeling intelligent agent datasets, and obtaining the centroid position of intelligent agents;

[0075] S5: Based on the features of the artificial road sign image and the centroid position of the agent, the angle is solved to obtain the offset angle and distortion angle.

[0076] The basic idea behind image angle estimation in this invention is to calculate the positions of artificial landmarks and intelligent agents relative to the image center, and simultaneously calculate the angle based on basic parameters such as camera intrinsics and focal length. The angle estimation workflow is as follows: Figure 3 As shown in the diagram, the functions of each part are described below:

[0077] Input image: Before acquiring the image, camera calibration is required to obtain intrinsic and distortion parameters, which provide basic parameters for subsequent image recognition and processing.

[0078] Image preprocessing: This part mainly includes image enhancement and binarization. The purpose of image enhancement is to increase the contrast between the target and the background, thereby improving recognition accuracy; the purpose of image binarization is to reduce computational complexity and highlight the outline of the target.

[0079] Artificial Road Sign Feature Extraction: This section primarily extracts features from artificial road sign images. First, the shape of the artificial road sign is designed based on rectangular and circular features. Then, the center position of the artificial road sign image is obtained through Hough circle transform and ellipse detection. Next, the corner information of the road sign is obtained through the Harris corner detection algorithm. Finally, the corner information is filtered using the center coordinates of the artificial road sign to obtain feature corners.

[0080] Intelligent Agent Target Recognition: This section primarily focuses on the recognition and extraction of intelligent agent images from complex indoor environments. An intelligent agent dataset is independently constructed and manually labeled. The centroid positions of the intelligent agents are obtained using the YOLO v3 target detection algorithm.

[0081] Angle estimation: This part mainly relies on the camera imaging model to solve for the angle and obtain the required offset angle and distortion angle.

[0082] In some embodiments, the distance and relative speed information between the intelligent agents are measured using acoustic technology.

[0083] In some embodiments, the sound information used to measure the distance and relative speed between intelligent agents is acquired through a sound sensor and a voice module, specifically including:

[0084] S6: The sound source base station periodically broadcasts sound signals into the environment, and the receiver collects the sound signals;

[0085] S7: Perform generalized cross-correlation on the acoustic signal, process the two hyperbolic frequency modulated signal components in the signal, and obtain the arrival time by using the fixed threshold method;

[0086] S8: Perform HFM signal composite on the arrival time to obtain the distance between agents and relative velocity.

[0087] The acoustic technology in this invention measures the sound information of the distance and relative motion speed between intelligent agents, which is acquired by a receiver. For example... Figure 4As shown, the sound source base station broadcasts near-ultrasonic signals into the environment at a certain period, and the receiver collects the sound signals to obtain the time of arrival (TOA) value. The near-ultrasonic signal used in this invention is a composite of hyperbolic frequency modulated signals from two different frequency bands within the range of 16kHz to 24kHz. Simultaneously, distance and relative velocity information are measured based on the composite hyperbolic frequency modulated signal.

[0088] The agent spacing and relative velocity are obtained through sound information. First, generalized cross-correlation (GCC) is performed on the sound signal. Then, the two hyperbolic frequency modulated (HFM) signal components in the signal are processed. Next, the time-of-arrival (TOA) is estimated using a fixed threshold method. Finally, the agent spacing and relative velocity are estimated by combining two HFM signals of different frequencies. The algorithm flowchart is shown below. Figure 5 As shown.

[0089] In some embodiments, the fusion of the acquired distance and relative motion speed information and angle information between intelligent agents to achieve cooperative localization among multiple intelligent agents specifically includes:

[0090] At time k, the agent's motion state is X. k ={x a,k ,v a,k x b,k ,v b,k} T The prediction equation and state update equation of the agent are as follows:

[0091]

[0092] Among them, F k Here is the state transition matrix.

[0093]

[0094] Y k For observations

[0095] Y k =[d k VR k v k θ 1，k θ 2，k ,β 1，k ,β 2，k α 1，k α 2，k ] T

[0096] h(X k ) is the observed value function; w k For predicting process noise; nk The noise in the observation process; the position x of agent a. a，k =(x a，k y a，k ) T The speed of agent a is v a，k =(v a，x，k v a，y，k ) T The position x of agent b b，k =(x b，k y b，k ) T The speed of agent b is v b，k =(v b，x，k v b，y，k ) T The sampling interval is Δt.

[0097] In the multi-agent cooperative localization process, to more accurately describe the motion state of the agents, the state-space model here utilizes a constant velocity (CV) motion model. At time k, the agent's motion state is X. k ={x a，k v a，k x b，k v b，k} T Where, the position x of agent a a，k =(x a，k y a，k ) T The speed of agent a is v a，k =(v a，x，k v a，y，k ) T The position x of agent b b，k =(x b，k y b，k ) T The speed of agent b is v b，k =(v b，x，k v b，y，k ) T The sampling interval is Δt, and the agent's prediction equation and state update equation are:

[0098]

[0099] Among them, F k Here is the state transition matrix.

[0100]

[0101] Y kFor observational values, in this invention, the distance d between two intelligent agents and the relative velocity vr between the intelligent agents can be obtained through sensors carried by the intelligent agents themselves. k The speed v of the intelligent agent's own movement k And the angles (θ1 and θ2, β1 and β2, α1 and α2) obtained by image processing of images acquired by the agent equipped with a camera, then Y k It can be represented as:

[0102] Y k =[d k VR k v k θ 1，k θ 2，k ,β 1，k ,β 2，k α 1，k α 2，k ] T

[0103] Where h(X) a，k ) is the observation function, w k To predict process noise, n k The noise during the observation process, and w k and n k Both follow a Gaussian distribution and are independent of each other, as stated below:

[0104] w k ~N(0,Q),n k ~N(0,R)

[0105] in

[0106] Finally, the prediction equation and state update equation of the agent are updated and solved based on the particle filter algorithm, so as to dynamically update the position of multiple agents and realize the cooperative localization among multiple agents.

[0107] This invention also discloses a multi-agent indoor collaborative positioning system based on artificial landmark and sound technology, comprising:

[0108] A setting unit is used to set up manual road signs;

[0109] Image acquisition unit, used to acquire images of artificial road signs and intelligent agents;

[0110] The analysis unit is used to analyze and process the acquired images of artificial road signs and intelligent agents to obtain angle information;

[0111] The distance and velocity acquisition unit is used to acquire distance and relative motion velocity information between intelligent agents;

[0112] The fusion unit is used to fuse the distance and relative motion speed information and angle information between the acquired agents to achieve cooperative localization among multiple agents.

[0113] The following simulation, using two-dimensional scene positioning as an example, describes the embodiments of the present invention.

[0114] When the two agents start moving, the cooperative localization algorithm is simulated for the system. The system contains two agents (denoted as a and b) and an artificial landmark. Velocity information is introduced into the particle filter to achieve localization.

[0115] Set the center point coordinates of the artificial road sign to (0, 0), the initial position coordinates of agent a to (-9, 17), define the motion of agent a as uniform motion at 1 m / s along the x-axis, the initial position coordinates of agent b to (13, 36), agent b also moves at a uniform speed of 1 m / s, and the positions of both are represented by red solid dots.

[0116] Based on experience, noise is added to the measurement data to simulate data changes in real-world scenarios. The angle measurement noise follows an R-order property. a ~N(0,10), distance measurement noise follows R l ~N(0,0.5), speed measurement noise follows R v ~N(0,0.1).

[0117] The obtained measurement coordinates are represented by solid blue dots. The simulation experiment involved 30 sampling iterations. The processed results after particle filtering are represented by hollow green circles. The positioning results are as follows: Figure 6 , Figure 7 As shown.

[0118] Figure 8 , Figure 9 The diagram above shows the CDF (Cooperative Localization Function) diagram for dynamic multi-agent cooperative localization. As can be seen from the diagram, under dynamic cooperative localization, agent a's localization error is 100% less than 0.65m, with a 90% probability of being less than 0.54m and an 80% probability of being less than 0.48m; agent b's localization error is 100% less than 0.82m, with a 90% probability of being less than 0.65m and an 80% probability of being less than 0.45m.

[0119] Therefore, the multi-agent cooperative localization method based on artificial landmarks proposed in this paper can also achieve high-precision position estimation in dynamic environments.

[0120] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0121] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0122] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0123] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0124] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A multi-agent indoor collaborative positioning method based on artificial landmarks and sound technology, characterized in that, include: Set up manual road signs (3); Acquire images of artificial road signs (3) and intelligent agents; The acquired images of the artificial road sign (3) and the intelligent agent are analyzed and processed to obtain angle information; the angle information includes the pose angle information between the intelligent agent and the artificial road sign (3), and the angle information between the intelligent agents; specifically including: S1: Camera calibration and input image, determining the basic parameters of the input image; S2: Perform image enhancement and binarization based on the basic parameters of the input image; S3: Extraction of image features of artificial road signs (3); S4: Recognition and extraction of intelligent agent images, autonomously constructing and labeling intelligent agent datasets, and obtaining the centroid position of intelligent agents; S5: Based on the image features of the artificial road sign (3) and the centroid position of the intelligent agent, the angle is solved to obtain the offset angle and distortion angle; Information on the distance and relative speed between intelligent agents is acquired; the audio information for this information is collected through a sound sensor and a voice module, specifically including: S6: The sound source base station periodically broadcasts sound signals into the environment, and the receiver collects the sound signals; S7: Perform generalized cross-correlation on the acoustic signal, process the two hyperbolic frequency modulated signal components in the signal, and obtain the arrival time by using the fixed threshold method; S8: Perform HFM signal composite on the arrival time to obtain the agent spacing and relative velocity. By fusing the acquired distance and relative motion speed information and angle information between intelligent agents, cooperative localization among multiple intelligent agents can be achieved; specifically including: exist At any given moment, the agent's motion state is The prediction equation and state update equation of the agent are as follows: in, Here is the state transition matrix. in, These are the observed values; It is a function of observed values; To predict process noise; Noise during the observation process; the position of agent a The speed of agent a is The position of agent b The speed of agent b is The sampling interval is .

2. The multi-agent indoor collaborative positioning method based on artificial landmarks and sound technology according to claim 1, characterized in that, The setting of artificial road signs creates a position triangle between the artificial road sign (3) and the intelligent agent.

3. The multi-agent indoor collaborative positioning method based on artificial landmarks and sound technology according to claim 1, characterized in that, The artificial road signs (3) and the images of the intelligent agents are acquired by mounting cameras on the intelligent agents.

4. The multi-agent indoor collaborative positioning method based on artificial landmarks and sound technology according to claim 1, characterized in that, The angles include: the offset angle of the artificial road sign (3) in the field of view of the agent, the offset angle of the agent in the field of view of another agent, and the image distortion angle of the artificial road sign (3).

5. A multi-agent indoor collaborative positioning system based on artificial landmarks and sound technology for implementing the multi-agent indoor collaborative positioning method based on artificial landmarks and sound technology as described in any one of claims 1 to 4, characterized in that, include: Setting unit, used to set up manual road signs (3); Image acquisition unit, used to acquire images of artificial road signs (3) and intelligent agents; The analysis unit is used to analyze and process the acquired artificial road signs (3) and intelligent agent images to obtain angle information; The distance and velocity acquisition unit is used to acquire distance and relative motion velocity information between intelligent agents; The fusion unit is used to fuse the distance and relative motion speed information and angle information between the acquired agents to achieve cooperative localization among multiple agents.

Images