A human-robot collaborative system, method, and storage medium

Abstract

Embodiments of the present application provide a human-robot collaborative working system and method, and a storage medium, and relate to the technical field of robots. The system comprises a user device and a robot. When communicating with the robot, the user device acquires a latest explored map of the robot and fuses the latest explored map with an existing map, acquires a temporary terminal point determined based on the fused map and plans a shortest path to the temporary terminal point, and guides the user to travel according to the planned path. The first robot acquires a plurality of first return combinations, generates a first target combination containing an earliest final arrival time and a corresponding communication time, and a return event. When a trigger condition of the return event is met in the exploration process, the first robot returns to the communication point in the return event to send the latest explored environment map to the user device when communicating with the user device. Through the present scheme, personnel can be safely and quickly guided to the terminal position in a communication-limited and unknown environment, and the task response efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of robotics, and in particular to a human-machine collaborative work system, method, and storage medium. Background Technology

[0002] When performing tasks in scenarios with limited communication and unknown environments (such as underground cave exploration, post-disaster search and rescue, etc.), robots are usually preferred to replace personnel for environmental exploration and safety checks.

[0003] There is already a great deal of research and application regarding multi-robot collaborative exploration. However, existing multi-robot collaborative exploration methods typically focus on maximizing environmental information acquisition to achieve maximum map coverage, failing to effectively address the demands of real-world tasks for rapid response. In other words, in task-oriented scenarios, it is necessary to guide personnel to the target location as quickly as possible to perform tasks such as disaster relief and search and rescue, thereby achieving rapid response.

[0004] However, existing multi-robot collaborative exploration methods typically involve the robot starting from a starting point and exploring a target location pre-set by the operator. Once a complete environmental map containing the target location is found, the robot returns to the operator's user device to communicate and send the explored environmental map. This allows the user device to plan the shortest path to the target location based on the robot's map. Because the operator must wait in place until the robot finds the environmental map containing the target location and returns to communicate with the user device, significant time is wasted.

[0005] Therefore, how to safely and quickly guide personnel to the target location in scenarios with limited communication and unknown environment, in order to improve mission response efficiency, has become an urgent technical problem to be solved. Summary of the Invention

[0006] The purpose of this invention is to provide a human-machine collaborative work system, method, and storage medium to safely and quickly guide personnel to a target location in scenarios with limited communication and unknown environments, thereby improving task response efficiency. The specific technical solution is as follows:

[0007] In a first aspect, embodiments of this application provide a human-machine collaborative work system, including:

[0008] When communicating with each robot, the user equipment obtains the latest explored environmental map from the robot and merges it with the existing environmental map to obtain a temporary endpoint that is close to the endpoint and in a safe area based on the merged map. Based on the merged map, the user equipment plans the shortest path to the temporary endpoint and guides the user carrying the user equipment to travel according to the planned path.

[0009] A first robot obtains multiple first return combinations, each first return combination including a communication point in the user equipment's current travel path and a return point in the first robot's current exploration path; it calculates the communication time between the first robot and the user equipment after the first robot returns from the return point in each first return combination to the communication point in that first return combination, and predicts the final arrival time of the user equipment to the destination position based on the calculated communication time; it generates a return event containing the earliest first target combination with the final arrival time and the corresponding communication time; when the triggering condition of the return event is met during the exploration process, it returns to the communication point in the return event to send the latest explored environment map to the user equipment when communicating with the user equipment, and continues exploration based on the map fused by the user equipment after communication.

[0010] Optionally, when the first robot communicates with the user equipment, it acquires and stores the current position of the user equipment, the current travel path, and the fused map of the user equipment; and during the exploration process, it updates the stored current position of the user equipment according to the stored current travel path and the preset speed of the user equipment.

[0011] The first robot,

[0012] For each preset point on the first boundary line between feasible and unexplored areas in the existing environmental map, the first cost corresponding to the preset point is calculated based on the shortest path from the current position of the first robot to the preset point, the shortest path from the current position of the user device to the preset point stored therein, and the shortest path from the preset point to the endpoint.

[0013] The preset point with the minimum first cost is taken as the first boundary point;

[0014] The shortest path from the first robot's current position to the first boundary point is planned as the current exploration path, and the exploration continues according to the current exploration path.

[0015] Optionally, the system includes a robot;

[0016] The process of obtaining multiple first return combinations includes:

[0017] When a preset time interval is reached during the exploration process, multiple first return combinations are obtained;

[0018] When the triggering condition of the return event is met during the exploration process, returning to the communication point in the return event to send the latest explored environment map to the user equipment during communication with the user equipment includes:

[0019] During the exploration process, the system checks whether the triggering condition of the latest returned event is met at the first time interval. If the condition is met, the system returns to the communication point in the latest returned event so that the latest explored environment map can be sent to the user equipment when communicating with the user equipment.

[0020] Optionally, calculating the communication time between the first robot and the user equipment after the first robot returns from the return point in each first return combination to the communication point in that first return combination, and predicting the final arrival time of the user equipment to the destination position based on the calculated communication time, includes:

[0021] For each first return combination, predict the first arrival time of the first robot traveling to the return point in the first return combination according to the current exploration path and from the return point to the communication point in the first return combination; and, based on the stored current position of the user equipment, predict the second arrival time of the user equipment traveling to the communication point in the first return combination according to the current travel path.

[0022] Based on the existing environmental map of the first robot, predict the first travel time of the shortest path from the communication point in the first return combination to the destination location of the user equipment;

[0023] The latest of the first arrival time and the second arrival time is determined as the communication time, and the first travel time is delayed based on the determined communication time to obtain the final arrival time of the user equipment to the destination position.

[0024] Optionally, the system includes multiple robots, and the multiple robots have a preset circular communication sequence;

[0025] When each first robot reaches the trigger condition of the currently pending communication event stored in its own memory, it goes to the communication point in the currently pending communication event to communicate with the second robot indicated by the currently pending communication event, and marks the currently pending communication event as executed;

[0026] During the communication process between the first robot and the second robot, the first robot generates and stores the next communication event between the first robot and the second robot, and if the communication time corresponding to the first target combination determined during the communication process with the second robot is earlier than the communication time of the next communication between the first robot and the second robot, a return event containing the determined first target combination and the corresponding communication time is generated.

[0027] The second robot is the robot adjacent to the first robot according to the circular communication sequence; the communication event to be executed by any robot is the earliest unexecuted communication event stored in the communication time.

[0028] Optionally, during communication with the second robot, each first robot acquires the latest environmental map explored by the second robot, the current location of the user device stored by the second robot, and the communication events stored by the second robot itself.

[0029] The generation of the next communication event between the first robot and the second robot includes:

[0030] The obtained environmental map is merged with its own existing environmental map, and the actual current location of the user device is determined based on the obtained current location of the user device and the current location of the user device stored in its own database.

[0031] Based on the merged map and the current actual location of the user device, exploration target points are assigned to the first robot and the second robot;

[0032] Based on the communication events currently pending to be executed by the first robot and the second robot, and the exploration target points assigned to the first robot and the second robot respectively, determine the communication time and communication point for the next communication between the first robot and the second robot;

[0033] Generate a next communication event that includes the determined communication time and communication point for the next communication, and obtain the next communication event between the first robot and the second robot.

[0034] Optionally, assigning exploration target points to the first robot and the second robot based on the fused map and the current actual location of the user device includes:

[0035] For each preset point on the first boundary line between feasible and unexplored areas in the merged map, the exploration cost of the preset point relative to the first robot is calculated based on the shortest path from the communication point to the preset point in the communication event to be executed by the first robot, the shortest path from the current actual location of the user device to the preset point, and the shortest path from the preset point to the destination location. Similarly, the exploration cost of the preset point relative to the second robot is calculated based on the shortest path from the communication point to the preset point in the communication event to be executed by the second robot, the shortest path from the current actual location of the user device to the preset point, and the shortest path from the preset point to the destination location.

[0036] Based on the exploration cost of each preset point relative to the first robot and the exploration cost of each preset point relative to the second robot, exploration target points are assigned to the first robot and the second robot respectively from the preset points.

[0037] Optionally, determining the communication time and communication point for the next communication between the first robot and the second robot based on the current communication events to be executed by the first robot and the second robot, and the exploration target points assigned to the first robot and the second robot respectively, includes:

[0038] For the latter robot (the first robot and the second robot), plan the shortest path from its current position through the communication point in the communication event to be executed by the latter robot to the assigned exploration target point, which will be the current exploration path of the latter robot. Based on the current exploration path of the latter robot, predict the first position of the latter robot at a first specified time. The first specified time is the communication time in the communication event to be executed by the first robot (the first robot and the second robot).

[0039] Determine the communication point for the next communication between the first robot and the second robot from the shortest path from the communication point in the current communication event to be executed by the previous robot to the first position;

[0040] Based on the determined communication point for the next communication, predict the communication time for the next communication between the first robot and the second robot.

[0041] Optionally, predicting the communication time of the next communication between the first robot and the second robot based on the determined communication point of the next communication includes:

[0042] Plan the shortest path from the communication point in the current communication event to the determined communication point of the next communication, and use it as the current communication path of the previous robot. Based on the current exploration path and communication path of the previous robot, calculate the third arrival time of the previous robot from its current position through the communication point in the current communication event to the determined communication point of the next communication.

[0043] Plan the shortest path from the first position to the determined communication point for the next communication, and use it as the current communication path of the next robot. Based on the current exploration path and communication path of the next robot, calculate the fourth arrival time of the next robot from the current position through the first position to the determined communication point for the next communication.

[0044] The latest of the third and fourth arrival times is determined as the communication time for the next communication between the first robot and the second robot.

[0045] Optionally, the first robot is the robot preceding the second robot;

[0046] The first robot obtained several first return combinations, including:

[0047] During communication with the second robot, the first robot receives multiple first return combinations;

[0048] When the triggering condition of the return event is met during the exploration process, returning to the communication point in the return event to send the latest explored environment map to the user equipment during communication with the user equipment includes:

[0049] During the exploration process, the system checks whether the triggering condition of the latest returned event is met at the first time interval. If the condition is met, the system returns to the communication point in the latest returned event so that the latest explored environment map can be sent to the user equipment when communicating with the user equipment.

[0050] Optionally, if the first robot is the robot following the second robot, when the first robot communicates with the second robot, it obtains the current travel path of the user device and the current exploration path of the second robot stored in the second robot.

[0051] Based on the latest travel path and the current exploration path of the second robot, multiple second return combinations are obtained; wherein, each second return combination includes a communication point in the latest travel path and a return point in the current exploration path of the second robot;

[0052] For each second return combination, based on the fused map, the communication time between the second robot and the user equipment, as well as the final arrival time of the user equipment to the destination position are predicted.

[0053] If the communication time corresponding to the second target combination with the earliest final arrival time is earlier than the communication time of the next communication between the first robot and the second robot, a return event containing the second target combination and the corresponding communication time is generated, and the generated return event is sent to the second robot. This allows the second robot to return to the communication point in the latest return event when the triggering condition of the latest return event is met during the exploration process. In this way, the robot can send the latest explored environment map to the user equipment when communicating with the user equipment, and continue to explore based on the map fused by the user equipment after communication.

[0054] Optionally, after generating a return event, each first robot predicts the second position of the first robot and the second robot at the communication time in the generated return event;

[0055] From the shortest path between the communication point in the generated return event and the second location, determine the communication point for the next communication between the first robot and the second robot;

[0056] The shortest path from the communication point in the generated return event to the determined communication point of the next communication is planned as the current communication path of the first robot and the second robot, and the seventh arrival time of the first robot to reach the determined communication point of the next communication is predicted according to the current communication path. The shortest path from the second position to the determined communication point of the next communication is planned as the current communication path of the second robot, and the eighth arrival time of the second robot to reach the determined communication point of the next communication is predicted according to the current communication path.

[0057] The latest of the seventh and eighth arrival times is determined as the communication time for the next communication between the first robot and the second robot, and the stored next communication event between the first robot and the second robot is updated according to the determined communication time and the determined communication point of the next communication.

[0058] Optionally, before obtaining a temporary endpoint that is close to the endpoint location and in a safe area based on the fused map, the user equipment determines whether the endpoint location is in a safe area in the fused map.

[0059] The process of obtaining a temporary endpoint that is close to the endpoint location and within a safe area, determined based on the fused map, includes:

[0060] If the endpoint is not located in a safe area in the merged map, then for each preset point on the second boundary line between the safe area and the potential risk area in the feasible area of ​​the merged map, the second cost corresponding to the preset point is calculated based on the shortest path from the current location of the user equipment to the preset point and the shortest path from the preset point to the endpoint; the preset point with the smallest corresponding second cost is determined as the temporary endpoint.

[0061] If the endpoint is located in a safe area of ​​the merged map, the endpoint will be designated as a temporary endpoint.

[0062] Optionally, if the first robot detects that the triggering condition of the latest returned event is not met, it may detect whether the triggering condition of the currently pending communication event has been met.

[0063] If the triggering condition of the current communication event to be executed is met, then the shortest path from the current position of the first robot to the communication point in the current communication event to be executed is planned, and the robot proceeds to the communication point in the current communication event to be executed according to the shortest path.

[0064] If the triggering conditions for the currently pending communication event are not met, the exploration will continue based on the existing environment map.

[0065] Secondly, embodiments of this application provide a human-machine collaborative work method, applied to the human-machine collaborative work system described in the first aspect above, wherein the human-machine collaborative work system includes: user equipment and robot, and the method includes:

[0066] Through the user equipment, when communicating with each robot, the latest explored environmental map is obtained from the robot and merged with the existing environmental map to obtain a temporary endpoint that is close to the endpoint and in a safe area based on the merged map. Based on the merged map, the shortest path to the temporary endpoint is planned and the user carrying the user equipment is guided to move according to the planned path.

[0067] The system uses a first robot to obtain multiple first return combinations. Each first return combination includes a communication point in the user equipment's current travel path and a return point in the first robot's current exploration path. The system calculates the communication time between the first robot and the user equipment after the robot returns from the return point in each first return combination to the communication point in that first return combination. Based on the calculated communication time, the system predicts the final arrival time of the user equipment to the destination position. A return event is generated, containing the earliest first target combination with the final arrival time and its corresponding communication time. When the triggering condition of the return event is met during exploration, the system returns to the communication point in the return event to send the latest explored environmental map to the user equipment during communication. After communication, exploration continues based on the user equipment's merged map.

[0068] Thirdly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the method described in the second aspect above.

[0069] Fourthly, embodiments of this application provide a computer program product comprising executable instructions that, when executed on a computer, cause the computer to perform the method described in the second aspect above.

[0070] Beneficial effects of the embodiments of the present invention:

[0071] The solution provided in this invention allows the user equipment to update its existing environmental map based on the environmental map obtained from the first robot. It then plans the shortest path to a temporary endpoint located in a safe area and close to the final destination based on the latest environmental map, guiding the user carrying the equipment along the planned path. When the first robot reaches the trigger condition for a return event during exploration, it can communicate with the user equipment, enabling the user equipment to obtain the latest environmental map explored by the first robot and redetermine the temporary endpoint located in a safe area and close to the final destination based on the latest environmental map, thereby updating the current travel path. In this way, the user equipment can proceed along the current travel path during the robot's exploration, achieving a mechanism where the robot explores while the user equipment guides the user. This eliminates the need for the user to wait for the robot to return and communicate with the user equipment after exploring an environmental map containing the endpoint location, reducing time wasted waiting in place. Since the temporary endpoint is always in a safe area, the user's safety is ensured by following the path planned by the user equipment.

[0072] Furthermore, since each first return combination includes a communication point on the user device's current travel path and a return point on the first robot's current exploration path, the first target combination in each first return combination has the earliest final arrival time, and the return event is generated based on the first target combination. Therefore, when the triggering condition of the return event is met, the first robot returns to communicate with the user device, enabling the user device to guide the user to the destination location faster based on the existing environmental map.

[0073] In summary, this solution enables the safe and rapid guidance of personnel to the destination in scenarios with limited communication and unknown environments, thereby improving task response efficiency.

[0074] Of course, implementing any product or method of the present invention does not necessarily require achieving all of the advantages described above at the same time. Attached Figure Description

[0075] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other embodiments can be obtained based on these drawings.

[0076] Figure 1 A schematic diagram of the structure of a human-machine collaborative work system provided in this application embodiment;

[0077] Figure 2 A schematic diagram of a potential risk area provided for an embodiment of this application;

[0078] Figure 3 A schematic diagram illustrating a preset ring communication sequence provided in an embodiment of this application;

[0079] Figure 4 A flowchart illustrating the state transition of a robot provided in this application embodiment;

[0080] Figure 5 A flowchart for generating communication events between robots is provided as an embodiment of this application;

[0081] Figure 6 A schematic diagram of a single-robot escort scenario provided in an embodiment of this application;

[0082] Figure 7 This is a schematic diagram of a multi-robot escort scenario provided in an embodiment of this application. Detailed Implementation

[0083] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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 based on this application are within the scope of protection of the present invention.

[0084] See Figure 1 This application provides a human-machine collaborative system, including a user device 110 and a robot 120;

[0085] When communicating with each robot 120, the user equipment 110 obtains the latest explored environmental map from the robot 120 and merges it with the existing environmental map to obtain a temporary endpoint that is close to the endpoint and in a safe area based on the merged map. Based on the merged map, the user equipment 110 plans the shortest path to the temporary endpoint and guides the user carrying the user equipment 110 to move according to the planned path.

[0086] The first robot in the system obtains multiple first return combinations. Each first return combination includes a communication point in the current travel path of the user equipment 110 and a return point in the current exploration path of the first robot. The system calculates the communication time between the first robot and the user equipment 110 after the first robot returns from the return point in each first return combination to the communication point in that first return combination. Based on the calculated communication time, the system predicts the final arrival time of the user equipment 110 to the destination position. The system generates a return event containing the earliest first target combination with the final arrival time and the corresponding communication time.

[0087] When the triggering conditions of the return event are met during the exploration process, the system returns to the communication point in the return event to send the latest explored environment map to the user equipment 110 when communicating with the user equipment 110. After communication, the system continues to explore based on the map fused by the user equipment 110. The triggering conditions include reaching the return point in the return event, or the time of reaching the communication point in the return event from the current location within a first time range, which is a preset time range before the communication time in the return event.

[0088] In this embodiment, the user equipment is a mobile electronic device that can be manually operated, such as a laptop, tablet, or a robot capable of autonomous movement. For example, in practical applications, an operator carries the user equipment to communicate with the robot, and the user equipment moves as the operator moves.

[0089] Each robot in the system is equipped with environmental sensors (LiDAR, cameras, etc.), a computing unit, and a wireless ad hoc network module, enabling point-to-point / multi-hop communication directly without an external communication base station. Each robot possesses Simultaneous Localization and Mapping (SLAM) capabilities, allowing it to detect unknown areas ahead and identify obvious signs of danger (such as suspicious moving targets) through environmental sensors. The robots involved in this application embodiment can be of various types, such as tracked robots, humanoid robots, or wheeled unmanned vehicles; the specific type of robot is not limited in this application embodiment.

[0090] The human-machine collaborative work system provided in this application embodiment may include one or more robots. When the system includes multiple robots, the first robot may be any robot in the system, and the robot group is initially deployed around the operator to jointly face the unknown task environment (such as complex terrain environments such as caves or disaster area tunnels).

[0091] During the robot's exploration, the areas it cannot reach are considered unexplored areas (i.e., unknown areas), while the known areas discovered by the robot, excluding obstacles, are considered feasible areas. Since obstacles at the boundary between feasible and unknown areas may affect the safety of the feasible area—for example, if there are deep pits, pools of water, or cliffs at the boundary—the operator may be placed in a dangerous situation when reaching this boundary. Therefore, to improve operator safety during task execution, in this embodiment, the feasible area is divided into a safe area and a potential risk area. The potential risk area is the boundary between the feasible and unexplored areas. When planning the operator's path, the endpoint of the path does not exceed the boundary of the safe area to ensure the operator's personal safety.

[0092] For example, potential risk zones can be identified within an existing environmental map where the distance from the boundary line between the feasible and unexplored areas falls within a preset length range. This preset length could be 3 meters, 5 meters, or similar. Figure 2 As shown in the diagram, the dotted areas represent potential risk zones, and the diagonally filled areas represent obstacles within the known areas. The blank areas within the known areas represent safe zones; in other words, the known areas include safe zones, obstacles, and potential risk zones. The areas within the known areas excluding obstacles are feasible zones, which include both safe zones and potential risk zones. During the robot's exploration, areas that cannot be detected due to obstacles or limited detection distance are considered unknown areas in the diagram. These potential risk zones are the boundary areas extending from the boundary between the feasible and unknown areas into the feasible area with a radius of r, where r is the aforementioned preset length.

[0093] In communication-constrained environments, user equipment can initiate communication requests at predetermined time intervals. Due to the limited signal coverage, any robot that moves into the signal coverage area of ​​the user equipment can receive the communication request and establish a communication connection with it. At this point, the user equipment and the robot communicate. For example, the predetermined time interval could be 1 second or 2 seconds, etc. In communication-constrained environments, the signal coverage area could be 5 meters or 10 meters, etc.

[0094] Initially, the user device and each robot communicate at the starting point. At this time, the first robot can send the latest environmental map it has explored near the starting point to the user device. The user device does not have an environmental map initially. After obtaining the latest environmental map explored by the first robot, it determines a temporary endpoint based on the environmental map. The first robot then explores according to the environmental map.

[0095] Furthermore, the operator can pre-set the endpoint location in the user equipment. For example, in practical applications, this endpoint location is the world coordinates of the destination the current task requires to reach. During initial communication with the first robot, the user equipment sends the pre-set endpoint coordinates to the first robot. This allows the first robot, after the initial communication ends, to plan an exploration path based on the initial environment map and the received endpoint coordinates.

[0096] When the system contains one robot, the first robot periodically generates return events during its exploration process. When the system contains multiple robots, the first robot generates a return event when communicating with the second robot, which is adjacent in a pre-defined circular communication sequence. For clarity, the exploration process of the first robot will be described first, followed by an introduction to the generation methods of return events when the system contains one robot and when it contains multiple robots.

[0097] When communicating with the user device, the first robot acquires and stores the user device's current location, current travel path, and the merged map of the user device; and during the exploration process, it updates the stored current location of the user device based on the stored current travel path and the preset speed of the user device.

[0098] It is understandable that the first robot and the user device communicate at a pre-agreed communication point (i.e., the communication point in the return event). In practical applications, since the coordinates of the first robot and the user device will not completely coincide, the first robot can obtain the current position of the user device during the communication process to more accurately obtain the current position of the user device. For example, after the first robot and the user device agree on a communication point, the robot is considered to have reached the communication point when it reaches a predetermined distance within the agreed communication point. Correspondingly, the user device is considered to have reached the communication point when it reaches a predetermined distance within the agreed communication point. This predetermined distance range can be 1m or 2m, etc.

[0099] In communication-constrained scenarios, the first robot and the user device can only communicate over short distances; that is, they must be within a limited communication coverage area to communicate. Since the first robot and the user device are relatively close during communication, in another implementation, the robot's position during communication can be used as the user device's current position.

[0100] When the user device communicates with the first robot, after updating its existing environmental map based on the latest environmental map obtained by the first robot, the user device will replan its current travel path based on the updated environmental map. Furthermore, the user device sends its current position and the newly planned travel path to the first robot, enabling the first robot to infer the user device's position at each subsequent moment based on the obtained position and travel path during the exploration process. For example, in practical applications, the first robot can periodically update the user device's current position at predetermined time intervals during the exploration process; the predetermined time interval could be 1 second, 2 seconds, etc.

[0101] For example, each time the first robot communicates with the user device, it can store the user device's current location, current path, and current timestamp as path information locally. During exploration, the first robot can calculate the user device's current location based on the locally stored path information and a preset user device speed. After calculating the user device's current location, the first robot can update the stored path information, ensuring that it always stores the latest path information, thus enabling it to monitor the user device's current location in real time. For example, the preset user device speed could be 0.5 m / s or 1 m / s, etc.

[0102] The first robot is specifically used to continue exploration based on the map fused by the user equipment after communication, through the following steps A1-A3:

[0103] A1. For each preset point on the first boundary line between the feasible area and the unexplored area in the existing environment map, calculate the first cost corresponding to the preset point based on the shortest path from the current position of the first robot to the preset point, the shortest path from the current position of the stored user device to the preset point, and the shortest path from the preset point to the destination position.

[0104] For example, preset points can be selected from the first boundary line at preset length intervals, and a first cost corresponding to each preset point can be calculated. The preset length interval can be 5m or 10m, etc.

[0105] Based on existing environmental maps, we can adopt... The path planning algorithm calculates the shortest path from the current position of the first robot to the preset point, the shortest path from the current position of the stored user device to the preset point, and the shortest path from the preset point to the destination position. After calculating these three shortest paths, the first cost corresponding to the preset point can be determined by combining the lengths of these three shortest paths.

[0106] The shortest path from the current location of the user device to the preset point is calculated using a path planning algorithm, taking into account known obstacles and threats within the existing environmental map of the first robot. Following this shortest path, the user device can bypass all obstacles, threats, and unexplored areas to reach the preset point. The shortest path from the preset point to the destination is calculated assuming no additional unknown obstacles or threats in the unexplored areas. Following this shortest path, the user device can bypass all known obstacles and threats from the preset point to reach the destination.

[0107] For example, the first cost corresponding to each preset point can be calculated using the following formula:

[0108] ;

[0109] in, This represents the first cost corresponding to the i-th preset point. This represents the length of the shortest path from the current position of the first robot to the preset point. This indicates the length of the shortest path from the user device's current location to the preset point. This indicates the length of the shortest path from the preset point to the terminal location. and These are preset weights. In practical applications, and It can be configured by relevant technical personnel based on experience. In one implementation method, and They are the same, that is, both are 0.5.

[0110] A2, take the preset point with the minimum first cost as the first boundary point.

[0111] A3. Plan the shortest path from the robot's current position to the first boundary point as the current exploration path, and continue exploring according to the current exploration path.

[0112] After calculating the first cost corresponding to each preset point, the preset point with the minimum first cost can be taken as the first boundary point, which is the current exploration target point. The first robot plans the shortest path to the first boundary point locally and moves there, returning to step A1 at predetermined time intervals during the exploration process; that is, the first robot replans the exploration path every certain period of time. For example, the predetermined time interval can be 1 minute or 2 minutes, etc.

[0113] The following section describes how return events are generated when the system contains a robot.

[0114] If the system includes a robot, the first robot generates a return event through the following steps B1-B5:

[0115] B1, when a preset time interval is reached during the exploration process, multiple first return combinations are obtained; each first return combination includes a communication point in the current travel path of the user equipment and a return point in the current exploration path of the first robot.

[0116] For example, the preset time interval could be 1 minute or 2 minutes, etc. During the exploration process, whenever the preset time interval is reached, multiple first return combinations can be obtained, and a return event can be generated based on the obtained multiple first return combinations, thereby enabling the periodic generation of return events. This exploration process does not include a return process.

[0117] Each time the first robot communicates with the user device, it can store the user device's current location, travel path, and current timestamp as path information locally. During exploration, the first robot can calculate the user device's current position based on the locally stored path information and the preset user device speed. After calculating the user device's current position, it can determine the untraveled portion of the current travel path (i.e., the remaining path). Furthermore, the first robot can also determine the untraveled portion of the current exploration path based on its own current position. In this embodiment, each communication point in the current travel path is determined from the untraveled portion of the current travel path, and similarly, each return point in the current exploration path is determined from the untraveled portion of the current exploration path.

[0118] For example, an enumeration method can be used to select return points and communication points. For instance, return points can be selected at a first preset distance from the untraveled portion of the first robot's current exploration path, resulting in a first number of return points. Similarly, communication points can be selected at a second preset distance from the untraveled portion of the user device's current travel path, resulting in a second number of communication points. For example, if the untraveled portion of the current exploration path is 1 km long, the untraveled portion of the user device's current travel path is 500 m long, the first preset distance is 10 m, and the second preset distance is 5 m, then 100 communication points and 100 return points can be selected.

[0119] For example, by iterating through each return point and each communication point, a return combination can be constructed to obtain multiple first return combinations.

[0120] Understandably, the communication point between the user equipment and the first robot is selected from the user equipment's current travel path. This allows the first robot to find the operator and communicate with the user equipment on the user equipment's current travel path when it returns to the communication point during the exploration process. In this way, the user equipment only needs to move along the current travel path before communicating with the first robot, without having to go to other paths to communicate with the first robot, which can reduce unnecessary time consumption.

[0121] B2, for each first return combination, predicts the first arrival time of the first robot traveling to the return point in the first return combination according to the current exploration path and from the return point to the communication point in the first return combination, and predicts the second arrival time of the user equipment traveling to the communication point in the first return combination according to the stored current position of the user equipment.

[0122] For each first return combination, the shortest path from the return point in that combination to the communication point can be planned based on the existing environmental map. Based on this shortest path and a preset robot speed, the travel time of the first robot from the return point to the communication point can be calculated. Then, based on the preset robot speed, the time when the first robot arrives at the return point in the first return combination from its current position along the current exploration path can be calculated. The first arrival time is obtained by delaying the calculated travel time from the first robot's arrival time at the return point. For example, the preset robot speed could be 2 m / s, 3 m / s, etc.

[0123] Accordingly, based on the stored current location and current travel path of the user equipment, the total distance the user equipment travels to the communication point in the first return combination can be calculated. Based on this total distance and a preset user equipment speed, the travel time of the user equipment can be calculated; this travel time is the quotient of the total distance and the preset user equipment speed. Delaying the current time by this travel time yields the second arrival time.

[0124] B3, based on the existing environmental map of the first robot, predicts the first travel time of the shortest path from the communication point in the first return combination to the destination location of the user equipment.

[0125] In this embodiment, the first robot can first plan the shortest path from the communication point in the first return combination to the destination location based on the existing environmental map, assuming that there are no obstacles or threats in the unexplored areas. After obtaining the shortest path, the quotient of the length of the shortest path and the preset speed of the user equipment is used as the first travel time of the user equipment from the communication point in the first return combination to the destination location.

[0126] B4. The latest of the first and second arrival times is determined as the communication time, and the first travel time is delayed based on the determined communication time to obtain the final arrival time of the user equipment to the destination.

[0127] It is understandable that, since the first arrival time represents the arrival time of the first robot at the communication point in the first return combination, and the second arrival time represents the arrival time of the user equipment at the communication point in the first return combination, the latest of the two times represents the time when both the user equipment and the first robot have arrived at the communication point and can communicate. Therefore, the latest of the first and second arrival times is the communication time. Because the communication speed between robots is very fast, the communication duration is negligible in this embodiment. Therefore, by delaying the communication time by the first travel time, the final arrival time of the user equipment at the destination position can be obtained.

[0128] B5: Select the first return combination with the earliest final arrival time from each first return combination as the first target combination, and generate a return event containing the first target combination and the corresponding communication time.

[0129] In this embodiment, after obtaining the final arrival time corresponding to each first return combination, the first return combination with the earliest corresponding final arrival time is determined as the first target combination. Then, a return event is generated that includes the return point and communication point in the first target combination, as well as the communication time corresponding to the first target combination.

[0130] It is understandable that, since the return event is generated based on the first target combination, and the first target combination has the earliest final arrival time among all the first return combinations, when the triggering condition of the return event is met, the first robot returns to communicate with the user equipment, enabling the user equipment to reach the destination position as early as possible based on the existing environmental map.

[0131] The triggering conditions for the return event include the time when the return point indicated by the return event is reached or the communication point of the return event is reached from the current position within a first time range. The first time range is a preset time range before the communication time in the return event. For example, the preset time range can be 3 seconds or 5 seconds, etc.

[0132] It is understandable that since the return event is determined based on the first target combination in each first return combination, and the return point in each first return combination is selected on the exploration path, which is the path leading to the boundary of the unexplored area, this exploration path includes paths within the potential risk zone. If the selected communication point is within the potential risk zone, the robot may not be able to reach the return point due to the threats present there, or the robot may not have enough time to communicate by the time it reaches the return point. Therefore, in this embodiment, it is also possible to detect whether the time of arrival at the communication point of the return event from the current position is within the first time range. If so, the return event is executed. It is understandable that by controlling the first robot to reach the communication point within the first time range, the first robot does not need to spend too much time waiting for the user to reach the communication point, thus making full use of the time for exploration.

[0133] For example, the first robot can plan the shortest path from its current position to the communication point in the return event based on its existing environmental map. Based on the length of the shortest path and the preset robot speed, the time when the first robot arrives at the communication point in the return event can be calculated.

[0134] The following section describes how return events are generated when the system contains multiple robots.

[0135] Since the system includes multiple robots, the return event of the first robot is determined when it communicates with the adjacent robot in the preset circular communication sequence. When two robots communicate, the next communication event between the two robots is planned first. After the next communication event is initially determined, it is decided whether to insert a return event before the next communication event. Therefore, in order to make the scheme layout clear, the method of determining the next communication event between the two robots will be introduced first, and then the method of generating the return event will be introduced.

[0136] If the system includes multiple robots, these robots have a preset circular communication sequence.

[0137] like Figure 3 The diagram illustrates a preset circular communication sequence. Robots i, j, k, l, and m communicate according to this circular communication sequence. Each first robot communicates only with its adjacent robot in the circular communication sequence. In practical applications, it can be pre-specified that robots i and j communicate first. Therefore, the communication order of each robot in each round of communication is: i communicates with j, j communicates with k, k communicates with l, l communicates with m, and m communicates with i.

[0138] When each first robot reaches the trigger condition of its stored current pending communication event, it proceeds to the communication point in the current pending communication event to communicate with the second robot indicated by the current pending communication event, and marks the current pending communication event as executed. The second robot is the robot adjacent to the first robot in the circular communication sequence. The trigger condition of the current pending communication event is the time when the robot arrives at the communication point in the communication event from its current position within a second time range. The second time range is a preset time range before the communication time in the communication event. The current pending communication event of any robot is the earliest unexecuted communication event stored in the communication time list.

[0139] The second robot, which is adjacent to the first robot in the circular communication sequence, can be either the robot preceding or following the first robot in the circular communication sequence. For example... Figure 3 As shown, if the first robot is robot i, then the second robot can be robot j or robot m. If the first robot is robot j, then the second robot can be robot i or robot k. For example, the preset time range can be 3s or 5s, etc.

[0140] Since the first robot can communicate with both the preceding and following robots in the circular communication sequence, the communication events stored in the first robot's own memory can include communication events with the preceding robot and communication events with the following robot. The first robot can store each communication event in the currently pending communication queue according to the order of the communication times contained in each event. At this time, the currently pending communication event is the first communication event in the communication queue. When the triggering condition of the currently pending communication event is met, the first robot proceeds to the communication point in the currently pending communication event to execute the communication event. At this time, the communication event is marked as executed. After marking, the currently pending communication event is updated to the next communication event.

[0141] Understandably, by controlling the first robot to arrive at the communication point within the second time frame, the first robot can avoid spending too much time waiting for the second robot to arrive, thus making full use of the time for exploration.

[0142] For example, the first robot can plan the shortest path from its current position to the communication point in the current communication event based on its existing environmental map. Based on the length of the shortest path and the preset robot speed, the time when the first robot arrives at the communication point in the current communication event can be calculated.

[0143] When the first robot precedes the second robot, the first robot obtains multiple first return combinations during communication with the second robot. Furthermore, during exploration, it checks whether the triggering condition of the latest obtained return event is met at a first time interval. If met, it returns to the communication point in the latest obtained return event to send the newly explored environment map to the user device when communicating with the user device. For example, the first time interval could be 3 seconds or 5 seconds, etc.

[0144] During communication with the second robot, the first robot generates and stores the next communication event between the first robot and the second robot. If the communication time corresponding to the first target combination determined during the communication with the second robot is earlier than the communication time of the next communication between the first robot and the second robot, a return event containing the determined first target combination and the corresponding communication time is generated.

[0145] The method by which the first robot obtains multiple first return combinations and determines the first target combination in each first return combination can be referred to the relevant descriptions of steps B1-B5 above, and will not be repeated here.

[0146] During the communication process between the first robot and the second robot, the next communication event between the first robot and the second robot is generated through the following steps C1-C5:

[0147] C1 retrieves the latest environmental map explored by the second robot, the current location of the user device stored by the second robot, and the communication events stored by the second robot itself.

[0148] It is understandable that when the first robot and the adjacent second robot determine the next communication event, all the calculations can be completed by the first robot. In practical applications, it is reasonable to pre-designate either the preceding robot or the following robot as the first robot.

[0149] Since all calculations are performed by the first robot, when the second robot communicates with the first robot, it can send information such as the latest explored environment map, its current exploration path, the current location of the stored user device, and its stored communication events to the first robot. This allows the first robot to combine the information from itself and the second robot to plan the next communication event with the second robot.

[0150] C2 merges the obtained environmental map with its own existing environmental map, and determines the actual location of the user device based on the obtained current location of the user device and the current location of the user device stored in its own database.

[0151] In this embodiment, the first robot continuously updates its existing environmental map based on the exploration results during the exploration process. When communicating with the second robot, the first robot merges the latest environmental map explored by the second robot with its own existing environmental map to update its own existing environmental map.

[0152] Furthermore, based on the current location of the user equipment stored by the second robot and the current location of the user equipment stored by the robot itself, the actual current location of the user equipment can be determined.

[0153] Understandably, due to the preset circular communication sequence among multiple robots, after the first robot communicates with the next robot, the next robot needs to execute a communication event with the robot immediately following it. At this point, the first robot has just finished executing a communication event with the robot preceding it, thus giving it more time to execute the return event. Therefore, in this embodiment, when two robots communicate, the robot in the preceding circular communication sequence always executes the return event.

[0154] Based on the above settings, if the second robot is the robot preceding the first robot in the circular communication sequence, then the current position of the user device stored in the second robot is taken as the current actual position of the user device; if the second robot is the robot following the first robot in the circular communication sequence, then the current position of the user device stored in the first robot is taken as the current actual position of the user device.

[0155] C3 assigns exploration target points to the first and second robots based on its own merged map and the current actual location of the user device.

[0156] For example, the first robot can calculate the shortest path from its current position to the preset point, the shortest path from the user device's current actual position to the preset point, and the total length of the shortest path from the preset point to the destination position for each preset point on the first boundary line between the feasible area and the unexplored area in the merged map, thereby obtaining the exploration cost of the preset point relative to the first robot. It can also calculate the shortest path from the second robot's current position to the preset point, the shortest path from the user device's current actual position to the preset point, and the total length of the shortest path from the preset point to the destination position, thereby obtaining the exploration cost of the preset point relative to the second robot.

[0157] After obtaining the exploration cost of each preset point relative to the first robot and the exploration cost of each preset point relative to the second robot, the point with the lowest exploration cost relative to the first robot can be used as the exploration target point of the first robot, and the point with the lowest exploration cost relative to the second robot can be used as the exploration target point of the second robot.

[0158] Alternatively, in one implementation, step C3 above is achieved through steps C31-C32:

[0159] C31, for each preset point on the first boundary line between the feasible area and the unexplored area in the merged map, calculates the exploration cost of the preset point relative to the first robot based on the shortest path from the communication point to the preset point in the communication event to be executed by the first robot, the shortest path from the current actual location of the user device to the preset point, and the shortest path from the preset point to the destination location. Also, calculates the exploration cost of the preset point relative to the second robot based on the shortest path from the communication point to the preset point in the communication event to be executed by the second robot, the shortest path from the current actual location of the user device to the preset point, and the shortest path from the preset point to the destination location.

[0160] In this embodiment, the method for calculating the exploration cost of the robot relative to any other robot can refer to the relevant description of calculating the first cost in step A1 above. The difference is that the current position of the robot in step A1 is replaced with the communication point in the communication event that the robot is currently to execute.

[0161] Understandably, if the first robot is the robot preceding the second robot in the preset circular communication sequence, then, to ensure that all robots always communicate according to the preset circular communication sequence (i.e., in each round of communication, the first robot communicates with the robot preceding it first, and then with the robot following it), the first robot must communicate with the robot preceding it before communicating with the second robot again. In other words, the first robot communicates with the robot preceding it before communicating with the second robot. Therefore, when calculating the exploration cost of each preset point relative to the first robot, the length of the shortest path between the first robot and the preset point from the communication point where it will communicate with the robot preceding it is considered. By selecting the preset point with the minimum exploration cost, the first robot can reach the exploration target point as quickly as possible from the communication point where it will communicate with the robot preceding it.

[0162] Accordingly, before the second robot can communicate with the first robot again, it must communicate with the robot following it. When calculating the exploration cost of each preset point relative to the second robot, the length of the shortest path between the communication point where the second robot communicates with the robot following it and the preset point must also be considered, so that the second robot can reach the exploration target point as quickly as possible from the communication point where it communicates with the robot following it.

[0163] Since the communication between the first robot and the second robot (communication with the second robot is the communication event being processed by the first robot) is the next communication event to be executed by the first robot, the communication point for the next communication between the first robot and the robot preceding it can be obtained based on the communication event to be executed by the first robot. Correspondingly, based on the communication event to be executed by the second robot, the communication point for the second robot to communicate with the robot following it can be obtained.

[0164] C32, based on the exploration cost of each preset point relative to the first robot and the exploration cost of each preset point relative to the second robot, assign exploration target points to the first robot and the second robot from the preset points.

[0165] In this embodiment, the preset point with the lowest exploration cost relative to the first robot can be determined as the exploration target point of the first robot, and the preset point with the lowest exploration cost relative to the second robot can be determined as the exploration target point of the second robot.

[0166] If the preset point with the lowest exploration cost is the same for the first robot and the second robot, then the preset point with the second lowest exploration cost can be assigned to the first robot or the second robot, so that the first robot and the second robot can be assigned different exploration target points.

[0167] In one implementation, the first robot may further consider the endpoint of the current exploration path when allocating exploration target points. The endpoint of any robot's current exploration path is the exploration target point previously allocated to that robot. For any robot, after calculating the exploration cost above, the exploration cost of each preset point relative to the robot is updated based on the distance from each preset point to the endpoint of the robot's current exploration path, so that each robot prioritizes exploration near the previously allocated exploration target point.

[0168] For example, for the exploration cost of a robot corresponding to each preset point, the distance from the preset point to the end of the robot's current exploration path can be used as a first coefficient, and the product or sum of the first coefficient and the exploration cost can be used as the new exploration cost. In this way, the greater the distance, the greater the updated exploration cost.

[0169] C4. Based on the communication events currently pending to be executed by the first robot and the communication events currently pending to be executed by the second robot, as well as the exploration target points assigned to the first robot and the second robot respectively, determine the communication time and communication point for the next communication between the first robot and the second robot.

[0170] C5 generates the next communication event, which includes the communication time and communication point of the next communication, thus obtaining the communication event between the first robot and the second robot.

[0171] After assigning exploration target points to the first and second robots, a communication path can be planned based on the existing environment map and the communication events currently pending between the two robots. Then, the communication point for the next communication is selected from the planned communication path. For example, a point can be randomly selected from the communication path as the next communication point, or an intermediate point on the communication path can be determined as the next communication point between the first and second robots. In practical applications, the communication cost corresponding to each point on the communication path can be calculated using enumeration, and the point with the minimum communication cost can be determined as the next communication point. This communication cost represents the total time required for the two robots to complete the next communication.

[0172] After determining the communication point for the next communication, the communication time can be predicted, thus enabling the generation of a next communication event containing the communication time and point. Upon receiving the next communication event with the second robot, the first robot can store the event itself and send it to the second robot, allowing the second robot to also store the event locally.

[0173] In one implementation, each first robot determines the communication time and communication point for the next communication with the second robot through the following steps C41-C43:

[0174] C41, for the latter of the first robot and the second robot, plan the shortest path from the current position of the latter robot through the communication point in the communication event to be executed by the latter robot to the assigned exploration target point, and use it as the current exploration path of the latter robot. Based on the current exploration path of the latter robot, predict the first position of the latter robot at a first specified time; the first specified time is: the communication time in the communication event to be executed by the first robot and the second robot.

[0175] C42, determine the communication point for the next communication between the first robot and the second robot from the shortest path from the communication point in the current communication event to be executed by the previous robot to the first position.

[0176] C43, based on the determined communication point for the next communication, predict the communication time for the next communication between the first robot and the second robot.

[0177] For example, the midpoint of the shortest path from the communication point to the first position in the current communication event to be executed by the previous robot can be selected as the communication point for the next communication. Then, based on the length of the shortest path and the preset robot speed, the total time for the robot to complete the shortest path can be calculated. Finally, by delaying the communication time in the current communication event to be executed by the previous robot by half of this total time, the communication time for the next communication between the first robot and the second robot can be determined.

[0178] Alternatively, in one implementation, step C43 is achieved through the following steps C431-C433:

[0179] C431: Plan the shortest path from the communication point in the current communication event to the determined communication point of the next communication, and use it as the current communication path of the previous robot. Based on the current exploration path and communication path of the previous robot, calculate the third arrival time of the previous robot from its current position through the communication point in the current communication event to the determined communication point of the next communication.

[0180] C432, plans the shortest path from the first position to the determined communication point for the next communication, and uses it as the current communication path for the next robot. Based on the current exploration path and communication path of the next robot, it calculates the fourth arrival time of the next robot from the current position through the first position to the determined communication point for the next communication.

[0181] C433, the latest of the third and fourth arrival times is determined as the communication time for the next communication between the first robot and the second robot.

[0182] For clarity, in the following text, the latter of the first and second robots will be referred to as the fourth robot, and the former of the first and second robots will be referred to as the third robot.

[0183] The communication event that the fourth robot is currently waiting to execute is a communication event with the robot following it. It is understandable that when the third robot communicates with the fourth robot, the next communication event between the third and fourth robots is planned. Furthermore, when planning the next communication event between the third and fourth robots, the next communication event between the third robot and the robot preceding it (i.e., the communication event that the third robot is currently waiting to execute), as well as the communication event between the fourth robot and the robot following it, need to be considered.

[0184] In other words, since the fourth robot needs to communicate with its successor robot after communicating with the third robot, the exploration path is planned using the shortest path from its current position through the communication point indicated by the currently pending communication event to the assigned exploration target point. Then, based on the preset robot speed, the position reached by the fourth robot following this exploration path at the next communication time between the third robot and its predecessor robot can be predicted, thus obtaining the first position.

[0185] After obtaining the first position, the shortest path from the communication point indicated by the current communication event to be executed by the third robot to the first position can be planned based on the existing environmental map, and this path can then be used as the communication path. The next communication point between the first and second robots can then be determined from this communication path.

[0186] For example, a point can be randomly selected from the communication path as the communication point for the next communication, or an intermediate point on the communication path can be determined as the communication point for the next communication between the first robot and the second robot. In practical applications, the communication cost corresponding to each point on the communication path can be calculated by enumeration, and the point with the minimum communication cost can be determined as the communication point for the next communication. This communication cost represents the total time taken for the two robots to complete the next communication.

[0187] For example, if the system includes four robots (robots i, j, k, and m), and the preset circular communication sequence is: i communicates with j, j communicates with k, k communicates with m, and m communicates with i, then, taking the communication between robot k (the third robot) and robot m (the fourth robot) as an example, during the exploration process, each robot locally determines the time to reach the communication point and arrives at the communication point according to the agreed communication time. At this time, robot k communicates with robot m. Then, robot m will obtain the next communication event from robot k.

[0188] Based on the communication event that requires the next communication with robot i, and the currently assigned exploration target point, robot m can plan the shortest path from its current position to the assigned exploration target point via the communication point indicated by the communication event with robot i. This path serves as the exploration path. Based on this exploration path, the position of robot m when robot k communicates with robot j (i.e., the first position) can be predicted. The communication point for the next communication between robot k and robot m is determined from the communication point of the next communication between robot k and robot j and the shortest path from the first position.

[0189] The shortest path from robot k to the communication point where it will communicate with robot j next, and then to the communication point where it will communicate with robot m next, is robot k's current communication path. Robot k's current exploration path is the shortest path from its current position to the communication point where it will communicate with robot j next (determined during its previous communication with robot j). Based on robot k's current exploration path and communication path, the complete path from its current position, passing through the communication point where it will communicate with robot j next, and then to the communication point where it will communicate with robot m next can be obtained. Based on this path, the arrival time of robot k at the communication point where it will communicate with robot m next (i.e., the third arrival time) can be predicted. Correspondingly, the arrival time of robot m at the communication point where it will communicate with robot j next (i.e., the fourth arrival time) can also be predicted.

[0190] Since the third arrival time represents the arrival time of the first robot at the communication point for the next communication, and the fourth arrival time represents the arrival time of the second robot at the communication point for the next communication, the latest of these two times indicates that both the first and second robots have arrived at the communication point for the next communication, at which point they can communicate. Therefore, the latest of the three arrival times is the communication time for the next communication between the first and second robots. After obtaining the communication time and the communication point for the next communication, a next communication event containing both the communication time and the communication point can be generated.

[0191] After the first robot generates the next communication event, it needs to determine whether a return event needs to be inserted before the next communication event. If a return event needs to be inserted, the next communication event is updated according to the required return event. The following describes the update process for the next communication event, which includes the relevant content for generating the return event, that is, the method by which the first robot generates the return event when the system contains multiple robots.

[0192] When the first robot is the successor to the second robot, during communication with the second robot, the first robot obtains the current travel path of the user device and the current exploration path stored by the second robot, and uses the obtained information to generate a return event for the second robot. At this time, the first robot also performs the following steps D1-D3:

[0193] D1. Based on the latest travel path and the current exploration path of the second robot, obtain multiple second return combinations; wherein, each second return combination includes a communication point in the latest travel path and a return point in the current exploration path of the second robot.

[0194] The latest obtained travel path is the current travel path of the user equipment stored in the second robot.

[0195] It is understandable that, since when two robots communicate, the robot that is in the circular communication sequence always executes the return event, when the second robot is the robot that precedes the first robot, the current travel path of the user device stored by the second robot is the current actual travel path of the user device.

[0196] For example, a third number of communication points can be selected from the latest obtained travel path, and a fourth number of return points can be selected from the current exploration path of the second robot. Multiple second return combinations can be constructed based on the selected return points and communication points. The specific method for selecting return points and communication points to construct the second return combinations can refer to the method for selecting return points and communication points to construct the first return combinations when the system includes one robot, and will not be repeated here.

[0197] D2, for each second return combination, predicts the communication time between the second robot and the user equipment, as well as the final arrival time of the user equipment to the destination position, based on its own fused map.

[0198] For example, step D2 can be achieved by the following steps D21-D23:

[0199] D21, for each second return combination, predicts the fifth arrival time of the second robot traveling to the return point in the second return combination according to the current exploration path and from the return point to the communication point in the second return combination, and predicts the sixth arrival time of the user equipment traveling to the communication point in the second return combination according to the user equipment's current actual position.

[0200] D22, based on its own merged map, predicts the second travel time of the shortest path from the communication point in the second return combination to the destination location for the user equipment.

[0201] D23 determines the latest of the fifth and sixth arrival times as the communication time, and delays the determined communication time by the second travel duration to obtain the final arrival time of the user equipment to the destination.

[0202] The implementation of steps D21-D23 can be referred to the relevant descriptions of steps B2-B4 above, and will not be repeated here.

[0203] D3. If the communication time corresponding to the second target combination with the earliest final arrival time is earlier than the communication time of the next communication between the first robot and the second robot, a return event containing the second target combination and the corresponding communication time is generated and sent to the second robot. This allows the second robot to return to the communication point in the latest return event when the triggering condition of the latest return event is met during the exploration process. This enables the robot to send the latest explored environment map to the user device when communicating with the user device, and to continue exploring based on the map fused by the user device after communication.

[0204] Step D3 can be referred to the relevant description in B5 above, and will not be repeated here. After receiving the return event sent by the first robot, the second robot performs the same detection process as the first robot described above.

[0205] If a return event is generated during communication between the first robot and the second robot, the next communication event needs to be updated based on the return event. If no return event is generated, the next communication event is not processed.

[0206] Each first robot updates the next communication event through the following steps E1-E4:

[0207] E1 predicts the second position of the second robot (the first robot and the second robot) at the time of communication in the generated return event.

[0208] It is understandable that after generating a return event, the second position of the fourth robot at that communication time can be predicted based on the communication time in the return event. That is, based on the fourth robot's current exploration path, the currently pending communication event, and the preset robot speed, the position of the fourth robot at the communication time indicated by the return event can be calculated; this position is the second position. The specific method for calculating the second position can refer to the implementation method for calculating the first position described above, and will not be repeated here.

[0209] E2 determines the communication point for the next communication between the first robot and the second robot from the shortest path between the communication point and the second position in the generated return event.

[0210] After obtaining the communication point and second location indicated by the generated return event, the shortest path between the communication point and the second location can be planned based on the existing environment map. Then, the communication point for the next communication between the first robot and the second robot can be selected from the shortest path. For example, a point can be randomly selected from the shortest path as the communication point for the next communication between the first robot and the second robot, or the intermediate point on the shortest path can be determined as the communication point for the next communication; both are reasonable.

[0211] E3, plan the shortest path from the communication point in the generated return event to the determined communication point of the next communication, as the current communication path of the previous robot, and predict the seventh arrival time of the previous robot to the determined communication point of the next communication according to the current communication path, and plan the shortest path from the second position to the determined communication point of the next communication, as the current communication path of the next robot, and predict the eighth arrival time of the next robot to the determined communication point of the next communication according to the current communication path.

[0212] The implementation of step E3 can be referred to the relevant descriptions of steps C431-C432 above, and will not be repeated here.

[0213] E4 determines the latest of the seventh and eighth arrival times as the communication time for the next communication between the first and second robots, and updates the stored next communication event between the first and second robots based on the determined communication time and the determined communication point for the next communication.

[0214] That is, a next communication event is generated, which includes the communication point for the next communication determined by step E2 above and the communication time for the next communication determined by step E4 above. The next communication event generated in step C5 is replaced with the next communication event generated in step C5 to update the next communication event between the first robot and the second robot.

[0215] Because the system generates return events periodically when it contains one robot, and generates return events when the first robot communicates with the second robot when the system contains multiple robots, return events will be generated multiple times regardless of whether the system contains one or multiple robots.

[0216] Each robot, during its exploration, checks at a first time interval whether the triggering condition of the latest returned event is met. If met, it returns to the communication point in the latest returned event to send the newly explored environment map to the user device during communication. After communication, it continues exploration based on the map fused by the user device. It is understandable that in practical applications, for... Figure 3 For robot i, if it is agreed that the first robot among the two communicating robots is the first robot, then when it communicates with robot j, robot i is the first robot, and it generates its own return event. The latest return event it receives is the latest return event it generates. If it is agreed that the second robot among the two communicating robots is the first robot, then when it communicates with robot j, robot i is the second robot, and the latest return event it receives is the return event sent by robot j.

[0217] For example, in practical applications, the first robot can update the received return events and only store the latest returned event. Alternatively, it can store all received return events, and after receiving a new return event, mark the historical returned events as invalid events so that it only checks whether the triggering conditions of the latest returned event are met.

[0218] For example, when the triggering condition of a return event is met, the first robot can plan the shortest path from its current position to the communication point in the return event based on its existing environmental map, and proceed to the communication point according to the shortest path. After both the user device and the first robot reach the communication point, the user device communicates with the first robot again. At this time, the user device can obtain the latest environmental map explored by the first robot and merge it with the existing environmental map to update the existing environmental map. After updating the existing environmental map, the user device obtains a temporary destination determined based on the merged map and plans the shortest path to the temporary destination. This planned path is the user device's current travel path.

[0219] For example, after the current travel path is planned, the user device can display the merged map, and the current travel path is displayed in the merged map, so that the user can travel according to the path displayed by the user device, realizing the process of the user device guiding the user to travel according to the planned path. The embodiments of this application do not limit the method of the user device guiding the user to travel according to the planned path. For example, the user device can also guide the user to travel by voice broadcasting the current travel path.

[0220] Before obtaining a temporary endpoint that is close to the destination and within a safe area based on the merged map, the user equipment determines whether the destination location is within a safe area in the merged map. If the destination location is within a safe area in the merged map, the destination location is determined as the temporary endpoint. In other words, the user equipment directly plans the shortest path from the user equipment's current location to the destination location based on the merged map as the current travel path.

[0221] If the destination is not located in a safe area in the merged map, a temporary destination that is close to the destination and located in a safe area can be determined by the following steps: For each preset point on the second boundary line between the safe area and the potential risk area in the feasible area of ​​the merged map, calculate the second cost corresponding to the preset point based on the shortest path from the current location of the user device to the preset point and the shortest path from the preset point to the destination; and determine the preset point with the minimum corresponding second cost as the temporary destination.

[0222] For example, multiple preset points can be obtained by selecting preset points at preset length intervals along the second boundary line. The preset length interval can be 5m or 10m, etc. For each preset point, based on the fused map, the following is applied... (A heuristic search algorithm) or (An incremental path planning algorithm) plans the shortest path from the user device's current location to the preset point, and the shortest path from the preset point to the destination location. The shortest path between the preset point and the destination location is planned based on the existing environmental map, assuming that there are no obstacles or unknown risks in the unexplored area.

[0223] The shortest paths between each pair of points involved in the embodiments of this application can all be obtained through... or The shortest path between any two points is obtained using an equal path planning algorithm. This application does not limit the specific method used to plan the shortest path between any two points.

[0224] After obtaining the two shortest paths, the sum of the lengths of the two shortest paths, or the average of the lengths of the two shortest paths, can be used as the second cost corresponding to the preset point. Then, the preset point with the smallest corresponding second cost among all preset points is determined as the temporary endpoint.

[0225] Understandably, when the destination is not within a safe zone on the existing environmental map, the shortest path from the user's current location to the temporary destination is planned as the current travel path. Since the temporary destination is located at the boundary between the safe zone and the potential risk zone, it does not exceed the safe zone's boundaries. Therefore, by following the current travel path, the operator can ensure that they remain within the safe zone, thus guaranteeing their personal safety during the journey. Furthermore, since the temporary destination is the second preset point with the lowest cost corresponding to the second boundary line, the operator can minimize the travel distance while approaching the terminal location by following the current travel path.

[0226] To further ensure user safety, a point within the safe zone can be selected as a temporary endpoint from a preset distance range corresponding to the second minimum cost preset point on the second boundary line. This preset distance range can be 1m or 2m, etc. Figure 2 As shown, within a known area, the operator's current path has a temporary endpoint, which is the location of the current target point in the image. This temporary endpoint is within a safe area, ensuring that the operator always moves within the known safe area.

[0227] Alternatively, the temporary endpoint can be determined by: determining the directional line connecting the current location of the user equipment to the endpoint location, selecting a temporary endpoint from the safe area of ​​the user equipment's existing environmental map along this directional line, or selecting a point within the safe area that is within a predetermined distance from the directional line as the temporary endpoint. Both of these methods are reasonable.

[0228] When the system includes multiple robots, if the first robot determines that the triggering condition of the latest generated return event is not met, it checks whether the triggering condition of the currently pending communication event has been met. If the triggering condition of the currently pending communication event has been met, the robot plans the shortest path from its current position to the communication point in the currently pending communication event and proceeds to the communication point in the currently pending communication event according to the shortest path; if the triggering condition of the currently pending communication event has not been met, the robot continues to explore based on its existing environment map.

[0229] During the exploration process, the first robot checks whether the trigger condition for the return event has been met at a first time interval. If the trigger condition for the return event has not been met, it checks whether the trigger condition for the currently pending communication event in the stored communication events has been met. If the trigger condition for the currently pending communication event has not been met either, it continues to explore based on the existing environment map. Otherwise, it executes the currently pending communication event, that is, it plans the shortest path from the first robot's current position to the communication point indicated by the currently pending communication event, and after the communication ends, it follows the shortest path to the communication point indicated by the currently pending communication event.

[0230] It is understandable that, since the return event determined by the first and second robots during the communication process is executed by the first robot (i.e., the third robot), the second robot (i.e., the fourth robot) can be in an exploration state, or in a communication state (i.e., communicating with the robot following the fourth robot) during the execution of the return event by the first robot, or in an exploration state for a period of time and in a communication state for a period of time.

[0231] During the exploration process, the first robot can also check at predetermined intervals whether it has explored the environmental map containing the destination location, and whether the destination location is within the safe area. If so, it can directly return to communicate with the user device. After receiving the environmental map returned by the first robot, the user device can plan the shortest path to the destination location within the safe area based on the environmental map explored by the first robot, and can send a termination command to the first robot. After receiving the command, the first robot can end the exploration, at which point the robot's task is completed and it can enter standby mode.

[0232] Understandably, the user device can update its existing environmental map based on the environmental map obtained from the first robot, and plan the shortest path to a temporary endpoint that is within a safe zone and close to the final destination based on the latest environmental map. It then guides the user carrying the device along the planned path. When the first robot reaches the trigger condition for a return event during its exploration, it can communicate with the user device, allowing the user device to obtain the latest environmental map explored by the first robot and redetermine the temporary endpoint that is within a safe zone and close to the final destination, thus updating the current travel path. In this way, the user device can proceed along the current travel path during the robot's exploration, enabling a mechanism where the robot explores while the user device guides the user. This eliminates the need for the user to wait for the robot to return and communicate with the user device after exploring an environmental map containing the endpoint location, reducing wasted time due to waiting in place. Since the temporary endpoint is always within a safe zone, the user's safety is ensured by following the path planned by the user device.

[0233] Furthermore, since each first return combination includes a communication point on the user device's current travel path and a return point on the first robot's current exploration path, the first target combination among all first return combinations has the earliest final arrival time, and the return event is generated based on the first target combination. Therefore, when the triggering condition of the return event is met, the first robot returns to communicate with the user device, enabling the user device to guide the user to the destination location more quickly based on the existing environmental map. This further reduces the time it takes for the user device to reach the destination location, enabling safe and rapid guidance of personnel to the destination location in scenarios with limited communication and unknown environments, thus improving task response efficiency.

[0234] The robot safety escort system provided in this application embodiment will now be described with reference to a specific example.

[0235] To overcome the shortcomings of existing technologies in collaborative exploration and escort missions under communication constraints, this example proposes a robot-assisted escort scheme for a human operator. This scheme centers on the human operator, enabling human-robot collaboration in unknown and potentially threatening environments with communication limitations. Communication and information exchange between robots, and between the robot and the operator, are only possible when the communication signal strength is above a threshold.

[0236] In this scheme, when multiple robots exist in the system, they cooperate and divide tasks. Generally, robots autonomously explore unknown areas, promptly identifying potential threats and planning safe routes. Simultaneously, robots will return to rendezvous with the operator, updating the local map information on the human terminal (corresponding to the user device mentioned above), thereby expanding the operator's feasible area. After obtaining the new map information, the human terminal replans its route and continues forward. Each robot can switch between exploration, communication, and return states according to task requirements, balancing the needs of environmental exploration, threat detection, and timely communication. The ultimate goal is to minimize the time required for the operator to reach the target location while ensuring safety.

[0237] When multiple robots are present in the system, each robot has three working states: exploration, communication, and return, to support the escort mission. The system is also compatible with a single robot, in which case it only has exploration and return states. Robots communicate intermittently according to a fixed circular topology (corresponding to the predetermined circular communication sequence mentioned above). At the start of the mission, all robots communicate once in a given order, planning the time and location of the next communication, thus fully constructing the communication topology. The robot then enters the exploration state, which becomes its default state. In the exploration state, the robot autonomously travels to the unexplored frontier of the environment, performing map expansion and threat detection tasks. The robot uses a frontier-based exploration algorithm to select new target points, bypassing known obstacles and gradually expanding the explored area, extending the known area towards the target location as much as possible. In the communication state, the robot travels to the planned communication point to communicate with the designated robot and exchange local information, while planning the next communication event, including the communication time and location. In the return state, the robot pauses further progress into the unknown area and moves towards the human operator to rendezvous with them. The returning robot acts as a temporary "messenger," bringing back environmental maps, threat intelligence, and other data acquired during its previous exploration. It also retrieves the operator's current location (corresponding to the user device's current location in the previous section) and path (corresponding to the travel path in the previous section). When the returning robot approaches the operator and communication conditions are met, both parties exchange data and update their status. After completing communication or the return task, the robot can switch back to exploration mode and resume environmental exploration. Through the coordinated operation of these three states, the system can achieve continuous exploration and information transmission simultaneously under intermittent communication conditions. This ensures that the operator obtains environmental information in a timely manner while maximizing the use of robot time for exploration, thus improving task response efficiency.

[0238] This plan includes the following four aspects:

[0239] F. The portable terminal carried by the operator (hereinafter referred to as the human terminal) continuously performs the following steps during the forward movement:

[0240] F1: When the human successfully communicates with the robot, the local map (corresponding to the existing environment map mentioned above) is first merged with the robot's map (corresponding to the latest environment map explored by the robot mentioned above). The resulting new map is used as the new local map, and then the process proceeds to F2.

[0241] F2 delineates potential risk zones on the local map of the operator's device. These risk zones are areas where the operator might be affected by unexplored threats from unknown areas. The specific delineation method considers the boundary points of all unknown areas (corresponding to the unexplored areas mentioned above) and known areas. Assuming threats exist at all boundary points, the area affected by these hypothetical threats is considered a potential risk zone. For example... Figure 2 As shown, after excluding the potential risk area and obstacles from the known area, the remaining area is the area that is absolutely safe for the operator, i.e., the safe area.

[0242] F3 checks if the destination (corresponding to the destination location mentioned above) is within the safe zone. If the destination is within the safe zone, the shortest path to the destination is directly calculated and the task is completed upon arrival. If the destination is not within the current safe zone, proceed to F4.

[0243] F4: The user terminal generates boundary points based on the current safe zone and the potential risk zone (corresponding to the preset points on the second boundary line mentioned above).

[0244] In step F5, after determining the current safe zone, the operator selects a temporary target point (corresponding to the temporary endpoint mentioned above) on the local map. The selection method for this temporary target point is as follows: for each boundary point generated in step F4 on the local map, calculate the shortest path length from it to the operator's own location (i.e., the operator's location). Then, assuming there are no unknown threats or obstacles in the unknown area, calculate the distance from that boundary point to the endpoint. Add these two distances to obtain the cost function for each boundary point (corresponding to the second cost mentioned above). Then, select the boundary point with the lowest cost as the current temporary target point, and plan the shortest path from that temporary target point on the local map as the currently planned path.

[0245] In step F6, the human-machine interface transmits its own position, the current temporary target point, and the planned path to the robot currently communicating with it. Then, it moves towards the temporary target point along the planned path until another robot returns to the operator to communicate with the human-machine interface. At this point, the process returns to step F1.

[0246] G. The robot performs the following steps repeatedly during normal exploration:

[0247] In G1, the robot determines whether the destination is within a known and safe area during its journey (i.e., after dividing the potential risk zone using the map according to steps F2, the destination is within the safe zone), allowing the operator to directly proceed to the destination within the safe zone. If so, the robot considers the exploration of the destination complete and returns directly to the operator to transmit the destination information (i.e., the map containing the destination). The robot's task is complete, and it enters standby mode. Otherwise, it proceeds to G2.

[0248] G2: If the task involves multiple robots, execute steps H1 and K1 below in sequence to determine whether it is necessary to return to the operator for communication or return to the agreed communication location (corresponding to the communication point mentioned above) to communicate with other robots. If the task involves only a single robot, execute only step H1. If both steps are completed normally and no return or communication state is entered, proceed to G3.

[0249] like Figure 4 As shown, the robot determines its next state by following these steps:

[0250] S401, exploration state, exploring towards the boundary point;

[0251] Initially, the robot is in exploration mode by default. When returning to S401 in subsequent steps, if the robot is not in exploration mode, it will switch to exploration mode.

[0252] S402, Determine whether the upcoming return event requires you to return; if yes, proceed to step S403; otherwise, proceed to step S405.

[0253] When the conditions for triggering the return event are met, it is determined whether the upcoming return event requires the user to return.

[0254] S403, determine whether the return point has been reached; if yes, proceed to step S404; otherwise, proceed to step S401.

[0255] S404, Return status, complete communication with the operator; and return to step S401 after completion;

[0256] S405, determine whether it is necessary to go to the communication point for communication; if yes, proceed to step S406; if no, proceed to step S401.

[0257] S406, Communication Status: Proceed to the communication point to complete communication. Then return to step S401 after completion.

[0258] G3: The robot generates boundary points based on the boundaries between known and unknown areas in the current map (corresponding to the preset points on the first boundary line mentioned above).

[0259] G4, the robot calculates the cost of each boundary point based on the local map (corresponding to the first cost mentioned above), including: (1) the length of the shortest path from its current position to that boundary point. (2) Calculate the operator's current position based on the latest operator position stored locally (corresponding to the current position of the user equipment mentioned above), the planned path (corresponding to the current travel path mentioned above), and the timestamp at that time; (3) Considering the known obstacles and threats within the robot's local map, calculate the length of the shortest path for the operator to reach the boundary point by bypassing all obstacles, threats, and unknown areas. Note that since the boundary points generated by the robot are likely to be located within the potential risk zone of the current map, the possible impact of unknown threats will not be considered when estimating the cost here; (4) Assuming there are no additional unknown obstacles and threat areas in the unknown area, calculate the length of the shortest path from the boundary point to the destination by bypassing all known obstacles and threat areas. (5) The results obtained above are weighted and calculated to obtain the cost of each boundary point. ,in and The given parameters; (6) In the case of multiple robots, the priority of the boundary points assigned to the robot during communication also needs to be considered. Assume that the boundary points assigned to the robot in the last communication were Then, for each boundary point within a certain range near the boundary point, based on their distance... The distance is reduced by the cost calculated in (4). (That is, the greater the distance, the less the cost is reduced; the closer the distance, the greater the cost is reduced.) The goal is to make the robot tend to prioritize [locations where the distance is greater than the distance between the two locations]. Explore a certain area around the robot. Note that if it's a single robot, this step can be skipped. After completing the above calculations, the robot obtains the cost of all boundary points and proceeds to G5.

[0260] In G5, the robot selects the point with the lowest cost among all boundary points as the current exploration target point, plans the shortest path to that point locally (corresponding to the exploration path mentioned above), and travels there. During the exploration process, it returns to step G1 at a certain frequency to re-loop.

[0261] H. When the robot is in the return state, it will return to the human end and perform the following steps:

[0262] In H1, the robot performs optimization calculations during the exploration process G1. The ultimate goal of the optimization is to minimize the operator's time to reach the destination while ensuring their safety. The robot's currently planned exploration path is now known. and the operator's current path We need to consider the following different variables through enumeration: (1) The position where the robot starts and returns. (2) The location where the robot is expected to communicate with the operator. Operator from After departure, the operator can be found on the operator's path and communication can begin; (3) the communication time between the two. (4) The operator's waiting time at the communication point can be obtained from the constraints of the above variables. After obtaining the above variables, it is necessary to minimize the ideal time for the operator to reach the destination from the current location. This involves calculating the time it takes for the operator to reach the destination from the current location, bypassing known obstacles and threats (excluding unknown areas), under the current map conditions (corresponding to the first travel time mentioned above). Delaying the communication time by this duration yields the idealized estimate of the time required to complete the task. (Corresponding to the final arrival time mentioned above). Under different enumerated variables, choose... The smallest group is selected as the current planned communication event. After the above calculations and planning are completed, proceed to H2.

[0263] H2, if the robot has reached the position where it should return to the operator. Or, it is determined that returning at this time is necessary to fulfill the agreed-upon conditions. If the operator's position is reached, proceed to H3; otherwise, return to step G2 to continue the exploration.

[0264] H3, the robot immediately returns to the operator along the shortest path (based on the shortest path from its current location to the communication point with the human). During communication, the robot transmits its map to the human, allowing the human to update information and obtain information such as the human's current location, temporary target point, and local planned path.

[0265] H4. After confirming that the communication is complete, the robot returns to step G1 and re-enters the exploration state.

[0266] I. The robot performs the following steps while in communication mode:

[0267] I1: The robot locally stores the planned subsequent communication events in an ordered queue. For the first event in the communication event queue, if it is determined that the communication event needs to be completed at this time (corresponding to the triggering condition of the current pending communication event mentioned above), that is, if it can reach the communication location before the planned time if it departs at this time, then proceed to I2; otherwise, return to G2.

[0268] I2. Assume that robots i and j communicate with each other. They first exchange known information, including the operator's position, the operator's current temporary target point, the operator's planned path, and the local map. The map information of the two is then merged to form their new local map.

[0269] I3, in the ring communication topology, robot i is ahead of robot j, so robot i can send a communication request when it arrives near the communication location. After receiving the request near the communication location, robot j determines whether it is the next communication to be carried out. If so, it starts to calculate the communication event. The specific steps are as follows: (1) First, obtain the boundaries of the known and unknown areas based on the new map obtained by fusion; (2) Calculate the boundary points that robots j and i should explore next (corresponding to the exploration target points in the above text). First, calculate the cost of the boundary points on the map (corresponding to each preset point on the second boundary line in the above text) relative to the two robots (corresponding to the exploration cost in the above text). Taking robot j as an example, the calculation process is similar to G4 (1)~(5), but the starting point of robot j needs to be changed from the current position to the current communication position with robot i. The process of calculating the cost for robot i is similar. Then, select the boundary point with the smallest cost for each robot j and i as the allocation result; (3) Based on the allocated boundary points, plan the shortest path between the two boundary points, and at the same time, on this path The communication point for the next communication is planned above. The selection method is to minimize the time it takes for both robots to reach the location after completing the exploration of the boundary. (4) Based on the planned boundary point and communication event, plan the specific exploration path for the two robots.

[0270] In step I4, after completing the planning in step I3, the two robots further optimize the next communication event. Here, the first consideration is whether information exchange with the human operator is necessary before the next communication. The calculation method is similar to H1. If the optimization result indicates that the two robots need to switch back to their previous state before the next communication, the robot with the earlier position in the communication topology, i.e., robot i, will be designated to return, and its return path will be planned. At this point, the original communication event needs to be optimized (i.e., updated) to obtain the new communication time and location. The optimization process is similar to H1, but the endpoints of the planned path are replaced with the boundary points assigned to robot j and the position where robot i communicates with the operator. If it is determined that a return is not necessary, no processing is performed on the original communication event.

[0271] After the calculation is completed, robot j, which is responsible for the calculation, passes the result to robot i, and the two robots return to G2 to continue the exploration.

[0272] See Figure 5 In a multi-machine collaborative solution, the following steps are performed:

[0273] S501, the robot arrives at the agreed coordinates at the specified time, forming a ring communication topology.

[0274] S502, robots communicate with each other, exchanging environmental maps, locations and other information; for example, robot i and robot j exchange information, including environmental maps, locations, exploration paths and stored human terminal information.

[0275] S503, plan the exploration route and agree on the time and place for the next communication between the two parties.

[0276] S504, determine whether it is necessary to return to the human end before the next communication; if yes, proceed to step S507, otherwise proceed to step S505.

[0277] S505: Each robot conducts distributed exploration according to the planned path.

[0278] S506, determine whether the exploration of the destination is complete; if yes, end; otherwise, return to step S501.

[0279] S507, replan the next communication event, inserting the human-end communication event before the next communication event;

[0280] S508, the robot communicates with the human terminal; after the communication ends, return to step S501.

[0281] The following sections will introduce this solution for single-robot escort scenarios and multi-robot escort scenarios respectively.

[0282] (a) Single-robot escort scenario;

[0283] In escort missions involving a single robot, this solution effectively ensures operator safety and improves mission efficiency. For example, in a cave environment containing potential threats, an autonomous mobile robot is deployed to collaborate with an operator carrying electronic devices. The operator's goal is to safely reach a designated location deep within the environment. In this environment, there is no ubiquitous communication signal, and due to terrain limitations, communication is impossible when the distance between the robot and the operator, or between robots themselves, is significant. Therefore, communication between robots and between the operator and the operator can only be temporary, via a wireless ad hoc network. In this scenario, the overall mission flow is as follows: Figure 6 As shown.

[0284] Initially, neither the human device (i.e., the user equipment) nor the robot has any map data. The robot communicates with the human device near the starting point. After acquiring the initial environment map, the human device begins to plan a path according to the aforementioned strategy, while the robot leaves the operator and begins to autonomously explore the unknown area. Due to the limited communication range, the robot needs to return to the operator multiple times during the exploration process to provide new map information. At this point, the robot begins to generate boundary points on the map and calculate the cost function for each boundary point, selects the optimal boundary point for exploration, plans a path and explores according to the path, and continuously generates new boundary points based on the current map.

[0285] During exploration, the robot acquires its current temporary target point and local path (i.e., the travel path mentioned above) through communication with the human terminal. Therefore, it can infer the operator's current position and the position at each subsequent moment. In other words, based on the path originally planned by the human terminal, it can predict the operator's current path and thus determine the operator's position. During exploration, the robot continuously optimizes and calculates the event of returning to the operator using known information. If it determines that it should return to the operator, it switches from exploration state to return state. At this point, it no longer heads towards the boundary point but plans the shortest path from its current position to the expected communication position (corresponding to the communication point mentioned above), using this path to return to the human terminal and reunite with the operator. At this time, the robot and the human terminal share map, location, and path information. Figure 6 As shown in the small image on the right, the new map obtained by merging the original map from the human device with the map transmitted back by the robot contains more environmental information. Each time the human device receives new map information, it replans its path, that is, it redefines the current target point (corresponding to the temporary endpoint mentioned above). This current target point is located at the boundary between the safe zone and the potential risk zone. Figure 6 The area sandwiched between the dotted lines in the small image on the right is a potential risk zone. The robot receives a new planned path from the human operator, avoiding the newly marked danger zone and ensuring it remains within the safe zone. The robot then re-enters exploration mode to explore the map's boundary points. At this point, both have completed a return event. Timely information sharing minimizes the distance the operator needs to walk.

[0286] The robot continued exploring in this mode until it discovered a path leading to the destination within a safe area on the map. At this point, a return event was triggered. The robot then shared the map with the operator, completed its task, and entered standby mode. The operator, after acquiring the latest map, planned the shortest path to the destination on their local map and proceeded there, safely reaching the target location and completing the task. Even when the operator reached the destination, a large area of ​​the environment remained unexplored, but this did not affect task completion. Compared to traditional global exploration strategies, this approach is more efficient. The system does not need to waste time exploring irrelevant areas; instead, it concentrates limited communication and exploration resources on task-related areas, significantly reducing the time required to complete the task.

[0287] In a single-robot scenario, since there is only one robot, there is no inter-robot communication or coordination within the system; the robot only needs to switch between exploration and return states. The robot autonomously determines the next exploration direction and timing by executing steps G1-G5, and intelligently decides when to return and communicate with the operator through steps H1-H4. In actual operation, when the operator approaches the edge of the known environment and is about to enter an unknown area, the robot often returns in advance according to an optimized strategy to exchange information with the operator, enabling the operator to obtain a new map of the environment ahead and potential risk warnings before continuing. Therefore, even in the event of communication interruption, the operator will not lose awareness of the environment for an extended period. Throughout the entire process, a path planning strategy is employed: the operator terminal uses the latest received map information to plan a safe path in real time (e.g., using...). The algorithm calculates the shortest safe path. Once it finds that the path ahead is impassable due to a newly discovered obstacle or threat, it immediately stops and replans. The robot uses a similar shortest path algorithm on its local map to plan a route to the target front point or communication point, ensuring that the exploration or communication task is completed in the fastest and most reliable way.

[0288] (ii) Multi-robot escort scenario;

[0289] In scenarios involving multi-robot collaborative execution, this solution demonstrates higher efficiency and robustness through teamwork. For example... Figure 7 As shown, the operator carries a tablet terminal as a human-machine interface device, and four autonomous mobile ground robots (robot i, robot j, robot k, and robot m) are deployed simultaneously. The communication topology is shown in the figure. Figure 3 They work together to complete the escort mission. Initially, the operator and the robot are located in the starting area on one side of the corridor. Due to the complex and changeable environment and the lack of communication base station support, the robot and the operator can only communicate directly at close range. However, each robot is equipped with a wireless self-organizing network device, which can realize point-to-point direct communication or multi-hop communication through other robots as relays.

[0290] After the task begins, the four robots first conduct initial communication according to a circular topology, and then plan subsequent communication events in sequence to build a complete communication mechanism: Robot i and Robot j exchange local map information observed from their respective starting points, merge them to establish an initial environment map under a unified coordinate system, assign exploration tasks, and agree on the time and meeting point for the next communication (corresponding to the communication points mentioned above). Then Robot j and Robot k conduct similar communication, and plan the communication events between Robot j and Robot k after the communication events planned by Robot j and Robot i, and so on until a circular communication is completed. Subsequently, each robot begins its exploration according to its assigned exploration direction and the cost of the boundary points. The operator then determines a temporary target (i.e., the current target point) based on the initial map and begins to explore along the planned path (i.e., the current target point). Figure 7 (Proceeding according to the original planned path of the Chinese side).

[0291] Under normal circumstances, the four robots are in an exploration state, autonomously selecting forward target points to expand the map, and communicating in the order of the ring communication topology when the specified time is reached. Taking the communication between robot k and robot m as an example, the two robots determine the time to reach the communication point locally during their exploration according to their own exploration paths, and arrive at the communication location at the agreed time. At this time, robot k sends a communication request to robot m. At this time, the two robots first fuse the map and threat information they have already acquired, and then obtain the boundary points on the latest map. At this time, robot m obtains the next communication event and time of robot k, that is, the communication event with robot j, and calculates the boundary points that the two robots will explore next according to the steps in I3(2). At this time, robot m calculates the path of robot k when robot j and robot k complete the communication, as well as its own path, and obtains the time and position of robot k after completing the exploration and communication, as well as its own position at that time (corresponding to the first position in the above text). Robot m plans the shortest path between the two positions and selects a point on the path as the next communication point, so as to enable the two robots to complete communication as quickly as possible, and obtains the time of the next communication based on the planned path and the coordinates of the selected communication point.

[0292] At this point, it is determined whether the robot needs to return to the operator before the next communication time between robot k and robot m. If so, a position on robot k's path is selected to switch to the return state, and the return time is obtained. Based on this, the path for robot k to return to the operator from that point and the time for completing the return event are planned, and the time when the return event is completed can be obtained. (Corresponding to the communication time of the return event indication in the above text) and robot k in Location (Corresponding to the second position in the above text), now we plan the robot m in... The position of time and The shortest path between robots k and m is used as the next rendezvous path (corresponding to the communication path mentioned above), and the specific time and location of this communication are calculated following similar steps as before. At this point, the communication event is complete, and the two robots re-enter the exploration state after determining their next exploration plan, until the next communication is needed or they need to return to the human end. During the forward movement, the operator's strategy and the robots' exploration and return-to-human-end steps are exactly the same as in the single-robot escort scenario, and will not be repeated here.

[0293] In the above multi-robot collaborative implementation, the robot team can cooperate with each other through the communication and coordination mechanism of this solution and complete the task as quickly as possible while ensuring the safety of the operator. For example, there are multiple instances of robots returning alternately: when one robot needs to return to deliver information to the operator, another robot continues to stay ahead to perform exploration and later obtains the latest map from its teammate through inter-robot communication, thus ensuring that the exploration work can proceed almost uninterrupted.

[0294] Throughout the escort process, communication primarily relies on a combination of phased rendezvous between robots and direct return communication between robots and operators. This ring-topic communication method greatly improves the system's robustness and exploration efficiency in complex terrain, enabling each robot to sequentially transmit key information to the operator and synchronously share their exploration results even with limited communication range.

[0295] The cooperation between multiple robots allows operators to obtain more effective map information earlier and plan better routes based on the new map information. For example... Figure 7 As shown, when robot i returns to communicate with the human terminal, it can send back the latest explored environmental map to the human terminal. This map contains the merged map formed when the robots communicate in a circular communication sequence. When robot i returns to the human terminal and reunites with the operator, the human terminal can obtain a more detailed map than before. Figure 6 More information (i.e.) Figure 7 The original map on the human client is updated to a new map on the human client, with the area sandwiched between the dotted lines being a potential risk zone. This allows the human client to plan a path around the danger zone based on the map returned by robot i. Figure 7 (New planned path for the human terminal). When any robot returns, by combining the original planned path on the human terminal and the operator's preset travel speed, the path the operator has traveled can be predicted, and the operator's real-time position can be obtained.

[0296] All four robots followed the aforementioned path planning strategy during their collaborative work: each robot planned its exploration and communication / return routes using the shortest path within its respective sub-task scope, and adjusted its next goal and path based on the merged map after each communication. This ensured that the entire robot team consistently and efficiently progressed towards a common ultimate goal. Therefore, by intelligently planning the timing and sequence of communication and exploration activities, the robot team could significantly reduce the time operators spent waiting for information and avoid exploration stagnation caused by all robots returning simultaneously, achieving efficient and safe escort of operators under conditions of limited communication and unknown environment.

[0297] This solution enables continuous collaborative exploration by multiple robots in unknown environments with limited communication, providing timely critical information to operators and significantly improving task execution efficiency and personnel safety. Specifically, by introducing a potential risk zone model, the solution allows operators to proactively avoid potential threats in unknown areas, effectively ensuring human safety during exploration. By optimizing the timing and route of robot return communication, the solution balances the conflict between environmental exploration and information feedback, reducing unnecessary waiting time and significantly shortening the total time for operators to reach the target. Simultaneously, multiple robots work collaboratively alternately through an intermittent communication mechanism in a ring topology, safely guiding personnel to the target without requiring a complete environmental exploration, minimizing the resources and time required to complete the task while ensuring safety.

[0298] Corresponding to the above method embodiments, this application also provides a human-machine collaborative work method applied to the human-machine collaborative work system in the above embodiments. The human-machine collaborative work system includes: user equipment and robot, and the human-machine collaborative work method includes:

[0299] Through the user equipment, when communicating with each robot, the latest explored environmental map is obtained from the robot and merged with the existing environmental map to obtain a temporary endpoint that is close to the endpoint and in a safe area based on the merged map. Based on the merged map, the shortest path to the temporary endpoint is planned and the user carrying the user equipment is guided to move according to the planned path.

[0300] The system uses a first robot to obtain multiple first return combinations. Each first return combination includes a communication point in the user device's current travel path and a return point in the first robot's current exploration path. The system calculates the communication time between the robot and the user device after returning from the return point in each first return combination to the communication point in that first return combination. Based on the calculated communication time, the system predicts the final arrival time of the user device to the destination position. A return event is generated, containing the earliest first target combination with the final arrival time and its corresponding communication time. When the triggering condition of the return event is met during exploration, the system returns to the communication point in the return event to send the latest explored environmental map to the user device during communication. After communication, exploration continues based on the user device's merged map. The triggering condition includes arriving at the return point in the return event, or the time of arrival at the communication point in the return event from the current position within a first time range, where the first time range is a preset time range before the communication time in the return event.

[0301] In another embodiment of the present invention, a computer-readable storage medium is also provided, which stores a computer program that, when executed by a processor, implements the steps of the above-described human-machine collaborative work method.

[0302] In another embodiment of the present invention, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute the human-computer collaborative work method described in the above embodiments.

[0303] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state disk (SSD)).

[0304] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0305] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the method embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0306] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of protection of the present invention.

Claims

1. A human-machine collaborative work system, characterized in that, include: User equipment. When communicating with each robot, the latest explored environmental map is obtained from the robot and merged with the existing environmental map to obtain a temporary endpoint that is close to the endpoint and in a safe area based on the merged map. The shortest path to the temporary endpoint is planned based on the merged map, and the user carrying the user equipment is guided to move according to the planned path. First Robot Multiple first return combinations are obtained, each first return combination including a communication point in the current travel path of the user equipment and a return point in the current exploration path of the first robot; Calculate the communication time between the first robot and the user equipment after the first robot returns from the return point in each first return combination to the communication point in that first return combination; predict the final arrival time of the user equipment to the destination position based on the calculated communication time; generate a return event containing the earliest first target combination of the final arrival time and the corresponding communication time; When the triggering condition of the return event is met during the exploration process, the system returns to the communication point in the return event to send the latest explored environment map to the user equipment when communicating with the user equipment. After communication, the exploration continues based on the map fused by the user equipment.

2. The system according to claim 1, characterized in that, When the first robot communicates with the user equipment, it acquires and stores the current location, current travel path, and the merged map of the user equipment. Furthermore, during the exploration process, the stored current position of the user equipment is updated based on the stored current travel path and the preset user equipment speed; The first robot, For each preset point on the first boundary line between feasible and unexplored areas in the existing environmental map, the first cost corresponding to the preset point is calculated based on the shortest path from the current position of the first robot to the preset point, the shortest path from the current position of the user device to the preset point stored therein, and the shortest path from the preset point to the endpoint. The preset point with the minimum first cost is taken as the first boundary point; The shortest path from the first robot's current position to the first boundary point is planned as the current exploration path, and the exploration continues according to the current exploration path.

3. The system according to claim 1, characterized in that, The system includes a robot; The process of obtaining multiple first return combinations includes: When a preset time interval is reached during the exploration process, multiple first return combinations are obtained; When the triggering condition of the return event is met during the exploration process, returning to the communication point in the return event to send the latest explored environment map to the user equipment during communication with the user equipment includes: During the exploration process, the system checks whether the triggering condition of the latest returned event is met at the first time interval. If the condition is met, the system returns to the communication point in the latest returned event so that the latest explored environment map can be sent to the user equipment when communicating with the user equipment.

4. The system according to claim 2, characterized in that, Calculate the communication time between the first robot and the user equipment after the first robot returns from the return point in each first return combination to the communication point in that first return combination, and predict the final arrival time of the user equipment to the destination position based on the calculated communication time, including: For each first return combination, predict the first arrival time of the first robot traveling to the return point in the first return combination according to the current exploration path and from the return point to the communication point in the first return combination; and, based on the stored current position of the user equipment, predict the second arrival time of the user equipment traveling to the communication point in the first return combination according to the current travel path. Based on the existing environmental map of the first robot, predict the first travel time of the shortest path from the communication point in the first return combination to the destination location of the user equipment; The latest of the first arrival time and the second arrival time is determined as the communication time, and the first travel time is delayed based on the determined communication time to obtain the final arrival time of the user equipment to the destination position.

5. The system according to claim 1, characterized in that, The system includes multiple robots, and the robots have a preset circular communication sequence. When each first robot reaches the trigger condition of the currently pending communication event stored in its own memory, it goes to the communication point in the currently pending communication event to communicate with the second robot indicated by the currently pending communication event, and marks the currently pending communication event as executed; During the communication process between the first robot and the second robot, the first robot generates and stores the next communication event between the first robot and the second robot, and if the communication time corresponding to the first target combination determined during the communication process with the second robot is earlier than the communication time of the next communication between the first robot and the second robot, a return event containing the determined first target combination and the corresponding communication time is generated. The second robot is the robot adjacent to the first robot according to the circular communication sequence; the communication event to be executed by any robot is the earliest unexecuted communication event stored in the communication time.

6. The system according to claim 5, characterized in that, During communication with the second robot, each first robot acquires the latest environmental map explored by the second robot, the current location of the user device stored by the second robot, and the communication events stored by the second robot itself. The generation of the next communication event between the first robot and the second robot includes: The obtained environmental map is merged with its own existing environmental map, and the actual current location of the user device is determined based on the obtained current location of the user device and the current location of the user device stored in its own database. Based on the merged map and the current actual location of the user device, exploration target points are assigned to the first robot and the second robot; Based on the communication events currently pending to be executed by the first robot and the second robot, and the exploration target points assigned to the first robot and the second robot respectively, determine the communication time and communication point for the next communication between the first robot and the second robot; Generate a next communication event that includes the determined communication time and communication point for the next communication, and obtain the next communication event between the first robot and the second robot.

7. The system according to claim 6, characterized in that, The process of assigning exploration target points to the first robot and the second robot based on the fused map and the current actual location of the user device includes: For each preset point on the first boundary line between feasible and unexplored areas in the merged map, the exploration cost of the preset point relative to the first robot is calculated based on the shortest path from the communication point to the preset point in the communication event to be executed by the first robot, the shortest path from the current actual location of the user device to the preset point, and the shortest path from the preset point to the destination location. Similarly, the exploration cost of the preset point relative to the second robot is calculated based on the shortest path from the communication point to the preset point in the communication event to be executed by the second robot, the shortest path from the current actual location of the user device to the preset point, and the shortest path from the preset point to the destination location. Based on the exploration cost of each preset point relative to the first robot and the exploration cost of each preset point relative to the second robot, exploration target points are assigned to the first robot and the second robot respectively from the preset points.

8. The system according to claim 6, characterized in that, The step of determining the communication time and communication point for the next communication between the first robot and the second robot based on the current communication events to be executed by the first robot and the second robot, and the exploration target points assigned to the first robot and the second robot respectively, includes: For the latter robot (the first robot and the second robot), plan the shortest path from its current position through the communication point in the communication event to be executed by the latter robot to the assigned exploration target point, which will be the current exploration path of the latter robot. Based on the current exploration path of the latter robot, predict the first position of the latter robot at a first specified time. The first specified time is the communication time in the communication event to be executed by the first robot (the first robot and the second robot). Determine the communication point for the next communication between the first robot and the second robot from the shortest path from the communication point in the current communication event to be executed by the previous robot to the first position; Based on the determined communication point for the next communication, predict the communication time for the next communication between the first robot and the second robot.

9. The system according to claim 8, characterized in that, The step of predicting the communication time of the next communication between the first robot and the second robot based on the determined communication point of the next communication includes: Plan the shortest path from the communication point in the current communication event to the determined communication point of the next communication, and use it as the current communication path of the previous robot. Based on the current exploration path and communication path of the previous robot, calculate the third arrival time of the previous robot from its current position through the communication point in the current communication event to the determined communication point of the next communication. Plan the shortest path from the first position to the determined communication point for the next communication, and use it as the current communication path of the next robot. Based on the current exploration path and communication path of the next robot, calculate the fourth arrival time of the next robot from the current position through the first position to the determined communication point for the next communication. The latest of the third and fourth arrival times is determined as the communication time for the next communication between the first robot and the second robot.

10. The system according to claim 5, characterized in that, The first robot is the preceding robot of the second robot; The first robot obtained several first return combinations, including: During communication with the second robot, the first robot receives multiple first return combinations; When the triggering condition of the return event is met during the exploration process, returning to the communication point in the return event to send the latest explored environment map to the user equipment during communication with the user equipment includes: During the exploration process, the system checks whether the triggering condition of the latest returned event is met at the first time interval. If the condition is met, the system returns to the communication point in the latest returned event so that the latest explored environment map can be sent to the user equipment when communicating with the user equipment.

11. The system according to claim 6, characterized in that, When the first robot is the robot following the second robot, the first robot, when communicating with the second robot, obtains the current travel path of the user equipment and the current exploration path of the second robot stored in the second robot; Based on the latest travel path and the current exploration path of the second robot, multiple second return combinations are obtained; Each second return combination includes a communication point in the most recently obtained travel path and a return point in the current exploration path of the second robot; For each second return combination, based on the fused map, the communication time between the second robot and the user equipment, as well as the final arrival time of the user equipment to the destination position are predicted. If the communication time corresponding to the second target combination with the earliest final arrival time is earlier than the communication time of the next communication between the first robot and the second robot, a return event containing the second target combination and the corresponding communication time is generated, and the generated return event is sent to the second robot. This allows the second robot to return to the communication point in the latest return event when the triggering condition of the latest return event is met during the exploration process. In this way, the robot can send the latest explored environment map to the user equipment when communicating with the user equipment, and continue to explore based on the map fused by the user equipment after communication.

12. The system according to claim 10 or 11, characterized in that, After generating a return event, each first robot predicts the second position of the first robot and the second robot at the communication time in the generated return event; From the shortest path between the communication point in the generated return event and the second location, determine the communication point for the next communication between the first robot and the second robot; The shortest path from the communication point in the generated return event to the determined communication point of the next communication is planned as the current communication path of the first robot and the second robot, and the seventh arrival time of the first robot to reach the determined communication point of the next communication is predicted according to the current communication path. The shortest path from the second position to the determined communication point of the next communication is planned as the current communication path of the second robot, and the eighth arrival time of the second robot to reach the determined communication point of the next communication is predicted according to the current communication path. The latest of the seventh and eighth arrival times is determined as the communication time for the next communication between the first robot and the second robot, and the stored next communication event between the first robot and the second robot is updated according to the determined communication time and the determined communication point of the next communication.

13. The system according to any one of claims 1-11, characterized in that, Before obtaining a temporary endpoint that is close to the endpoint and in a safe area based on the merged map, the user equipment determines whether the endpoint location is in a safe area in the merged map. The process of obtaining a temporary endpoint that is close to the endpoint location and within a safe area, determined based on the fused map, includes: If the endpoint is not located in a safe area in the merged map, then for each preset point on the second boundary line between the safe area and the potential risk area in the feasible area of ​​the merged map, the second cost corresponding to the preset point is calculated based on the shortest path from the current location of the user equipment to the preset point and the shortest path from the preset point to the endpoint; the preset point with the smallest corresponding second cost is determined as the temporary endpoint. If the endpoint is located in a safe area of ​​the merged map, the endpoint will be designated as a temporary endpoint.

14. The system according to claim 10 or 11, characterized in that, If the first robot detects that the triggering condition of the latest returned event is not met, it checks whether the triggering condition of the currently pending communication event has been met. If the triggering condition of the current communication event to be executed is met, then the shortest path from the current position of the first robot to the communication point in the current communication event to be executed is planned, and the robot proceeds to the communication point in the current communication event to be executed according to the shortest path. If the triggering conditions for the currently pending communication event are not met, the exploration will continue based on the existing environment map.

15. A human-machine collaborative work method, characterized in that, Applied to the human-machine collaborative work system according to any one of claims 1-14, the human-machine collaborative work system comprising: user equipment and robot, the method comprising: Through the user equipment, when communicating with each robot, the latest explored environmental map is obtained from the robot and merged with the existing environmental map to obtain a temporary endpoint that is close to the endpoint and in a safe area based on the merged map. Based on the merged map, the shortest path to the temporary endpoint is planned and the user carrying the user equipment is guided to move according to the planned path. The system uses a first robot to obtain multiple first return combinations. Each first return combination includes a communication point in the user equipment's current travel path and a return point in the first robot's current exploration path. The system calculates the communication time between the first robot and the user equipment after the robot returns from the return point in each first return combination to the communication point in that first return combination. Based on the calculated communication time, the system predicts the final arrival time of the user equipment to the destination position. A return event is generated, containing the earliest first target combination with the final arrival time and its corresponding communication time. When the triggering condition of the return event is met during exploration, the system returns to the communication point in the return event to send the latest explored environmental map to the user equipment during communication. After communication, exploration continues based on the user equipment's merged map.

16. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method of claim 15.

17. A computer program product, characterized in that, The computer program product includes executable instructions that, when executed on a computer, cause the computer to perform the method of claim 15.

Images