A multi-machine cooperative decision control method and system for robot navigation

Abstract

The application discloses a kind of multi-machine cooperative decision control method and system for robot navigation, which comprises: determining at least one local machine from multiple machines, obtaining local navigation information, and receiving at least one cooperative information of other machine;According to the local navigation information and the cooperative information, the relative state relationship and the path association relationship between the local machine and any other machine are determined;Based on the relative state relationship, the path association relationship and the real-time control parameter, the traffic interaction relationship between the local machine and the other machine is identified to determine the conflict type in the multi-machine scene;According to the conflict type, define the conflict strength of multiple machines, obtain the environmental obstacle information to construct the local scene constraint, determine the cooperative traffic decision result based on the local machine in the current scene based on the conflict strength of multiple machines and the local scene constraint;According to the cooperative traffic decision result, generate constraint information for local planning and motion control, to control the local machine and the other machine to complete cooperative traffic in shared traffic space.

Description

Technical Field

[0001] This invention relates to the field of multi-machine collaborative technology, and in particular to a multi-machine collaborative decision control method and system for robot navigation. Background Technology

[0002] Existing mobile robots are commonly used in scenarios such as logistics and distribution, cleaning operations, park inspections, and garbage transfer. They typically employ a navigation method that combines global planning with local obstacle avoidance. However, in multi-robot scenarios, the dynamic obstacle relationships in local obstacle avoidance will change into complex interactions between multiple robots. Among these challenges, how to achieve safe and efficient multi-robot collaborative control has become a key technical issue.

[0003] In existing technologies, multi-machine cooperative navigation control has the following problems: Typically, non-machine objects are only considered obstacles, while the machine object performs reactive obstacle avoidance based on real-time position and velocity parameters. However, this method cannot effectively utilize the motion parameters of non-machine objects, resulting in insufficient effectiveness of global decision-making. Secondly, in multi-machine scenarios such as oncoming traffic, lane occupancy, intersection rushing, or multi-vehicle congestion, the lack of differentiated decision-making rules for individual machines can easily lead to problems such as individual machine accidental stopping, multi-machine oscillation, redundant planning, and local deadlock. Further multi-machine scenarios also present environmental constraints. If environmental information and multi-machine priority relationships are not combined, the impact of redundant planning and local deadlock on global decision-making will be amplified. At the same time, global multi-machine collaborative decision-making cannot be transformed into dynamic set constraints perceived by local planning, resulting in a serious disconnect between local control and multi-machine collaborative decision-making.

[0004] It is evident that existing technologies have shortcomings that urgently need to be addressed. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a multi-machine collaborative decision control method and system for robot navigation, which can realize safe and efficient operation of multi-machine collaborative control and improve safety, stability and operating efficiency in complex multi-machine scenarios.

[0006] To address the aforementioned technical problems, the first aspect of this invention discloses a multi-machine cooperative decision-making and control method for robot navigation, the method comprising: Identify at least one local machine from among multiple machines, obtain local navigation information, and receive cooperative information from at least one other machine; Based on local navigation information and cooperative information, determine the relative state relationship and path association between the local machine and any other machine; Based on relative state relationships, path association relationships, and real-time control parameters, the communication interaction relationships between the local machine and other machines are identified to determine the type of conflict in multi-machine scenarios. The multi-machine conflict intensity is defined based on the conflict type in the multi-machine scenario, and the local scenario constraints are constructed by acquiring environmental obstacle information. Based on the multi-machine conflict intensity and the local scenario constraints, the local cooperative passage decision result in the current scenario is determined. Based on the results of the collaborative passage decision, constraint information is generated for local planning and motion control to control the local machine and other machines to complete collaborative passage within the shared passage space.

[0007] As an optional implementation, in the first aspect of the present invention, the local navigation information includes the local machine's number, timestamp, map name, local path, target point, robot model identifier, task status, multi-machine status, scene information, and real-time control parameters. The collaborative information includes the other machine's ID, timestamp, map name, local path, target point, robot model identifier, task status, multi-machine status, scene information, and real-time control parameters.

[0008] The collaborative information is broadcast by any of the other machines and received by the local machine.

[0009] As an optional implementation, in the first aspect of the invention, determining the relative state relationship and path association relationship between the local machine and any other machine based on local navigation information and cooperative information includes: Calculate the longitudinal relative distance, lateral relative distance, and heading difference between the machine and any other machine based on the machine's attitude and the attitudes of other machines, and obtain the relative state relationship between the machine and all other machines. The target vector from the local position to the target point is calculated based on the local target point and local pose in the local navigation information. The relative vector from the local machine to the other machine is calculated based on the local target point and the other machine's position. The path judgment value of the local machine is obtained by performing a dot product between the target vector and the relative vector. The hybrid product is calculated based on the local pose, the other machine pose, and the target point, as well as the Euclidean distance between the local machine and the target point. The path planning value is obtained by the ratio of the hybrid product to the Euclidean distance. The path judgment value determines the front-to-back relationship of other machines relative to the machine in the forward direction, and the path planning value determines the lateral deviation of other machines from the machine's path. The path association relationship between the machine and any other machine is determined by the front-to-back relationship and the lateral deviation.

[0010] As an optional implementation, in the first aspect of the present invention, based on relative state relationships, path association relationships, and real-time control parameters, the communication interaction relationship between the local machine and other machines is identified to determine the conflict type in a multi-machine scenario, including: The geometric parameters of the robot and other robots are determined based on the robot model identifiers, and the robot geometric model is constructed. The geometric parameters include the robot's forward length, backward length, left width, and right width. The minimum safe distance is calculated based on the geometric parameters of the machine and other machines and the preset collision safety margin. The collision risk between the machine and other machines is identified by the relationship between the minimum safe distance and the relative state. Set interaction thresholds for corresponding conflict types for communication interaction relationships, and calculate the difference in direction angle between the other machine and the local machine. The conflict types include perpendicular intersection, same-direction lane occupation, and oncoming traffic meeting. When the absolute difference between the direction angle and the vertical angle is less than the first interaction threshold, the conflict type corresponding to the passage interaction relationship between the local machine and other machines is determined to be vertical intersection. When the difference in direction angles is less than the second interaction threshold and the longitudinal relative distance is less than the minimum safe distance, the conflict type corresponding to the passage interaction relationship between the local machine and other machines is determined as same-direction lane occupation. When the absolute difference between the direction angle and the horizontal angle is less than the third interaction threshold, and the path planning value is less than the minimum safe distance, the conflict type corresponding to the passage interaction relationship between the local machine and other machines is determined as oncoming traffic.

[0011] As an optional implementation, in the first aspect of the present invention, the multi-machine conflict intensity is defined according to the conflict type in the multi-machine scenario, and local scene constraints are constructed by acquiring environmental obstacle information. Based on the multi-machine conflict intensity and the local scene constraints, the local-based cooperative passage decision result in the current scenario is determined, including: The multi-machine conflict intensity is defined by a weighted sum of the number of robots corresponding to all conflict types and the weight of each conflict type. Obtain the obstacle point cloud in the local obstacle environment, and count and count the point cloud of each obstacle region based on the preset saturation upper limit to obtain the local scene constraints. The obstacle regions in the obstacle environment include the forward region, the left region, and the right region. The maximum allowable linear speed of the robot is determined based on the intensity of multi-machine conflict and the preset deadlock threshold and speed limit threshold. The robot's search replanning flag is determined based on the constraints of the local scene and the preset constraint threshold. The robot's multi-machine state factors are determined based on its own navigation information and collaborative information, and the task state factors of the robot and other robots are determined based on its own navigation information and collaborative information. For any robot, the decision priority is calculated using a priority function based on the maximum permissible linear velocity, the search replanning flag, the multi-machine state factor, and the task state factor. Based on decision priority, output passage decisions for robots in multi-robot scenarios, as well as collaborative passage decisions under all passage decisions of the robot itself.

[0012] As an optional implementation, in the first aspect of the invention, generating constraint information for local planning and motion control based on the cooperative passage decision results to control the local machine and other machines to complete cooperative passage within a shared passage space includes: Based on the geometric model of the other machine, generate contour vertices for local planning for the other machine in the case of collision and path conflict; The contour vertices are transformed to the global coordinate system on the other machine's pose and sampled along the contour boundary to form the other machine's global contour point set. The other machine's global contour point set is regarded as the dynamic obstacle boundary input to the local planning map as obstacle constraint information. The target local planning boundary line point set is generated based on the local pose and the pose of other local machines. The local boundary line point set serves as the boundary line constraint information in the local planning map. The target local machine is the key local machine in the vehicle meeting conflict and path conflict. Based on the real-time control parameters of the machine, the motion working domain under the longitudinal speed limit and curvature coordination control constraints is determined. Within the selected range of the motion working domain, the longitudinal speed limit based on the curvature constraint information of the motion trajectory is used to maintain curvature within the maximum allowable linear velocity. Based on obstacle constraint information, separator line constraint information, and curvature constraint information, continuous control is performed on the cooperative passage of the local machine and other machines within the shared passage space.

[0013] As an optional implementation, in the first aspect of the invention, generating a set of dividing line points for the local planning of the target machine based on the local machine pose and the other machine pose includes: Calculate the path orientation angle of the target machine based on the continuous poses of the target machine and other machines, and transform the poses of other machines into the path coordinate system of the target machine based on the path orientation angle to obtain the forward distance and lateral offset distance of other machines relative to the target machine. The effective conditions for generating the separator line range are preset, and the safety conditions for the separator line width are determined based on the geometric models of the local machine and other machines. Under the effective conditions and safety conditions, the slope of the separator line point set is calculated based on the forward distance and the lateral offset distance. The lateral coordinate constraints of the dividing line point set are determined based on the classification of the lateral position and lateral offset distance of other machines relative to the local machine. Generate the set of dividing line points for the target machine under local planning based on slope and lateral coordinate constraints.

[0014] As an optional implementation, in the first aspect of the present invention, the motion working domain corresponding to the real-time control parameters of the machine is determined by comparing the linear velocity and angular velocity in the real-time control parameters with a preset minimum linear velocity threshold and a rotational state angular velocity threshold. The curvature constraint of the motion trajectory is defined according to the motion working domain, where the curvature is the rate of change of heading per unit path length when the machine travels along the current motion trajectory; The longitudinal linear velocity is limited and adjusted based on the maximum permissible linear velocity in the collaborative passage decision results, and the angular velocity in the actual control parameters is corrected based on the longitudinal linear velocity after the limit adjustment according to the curvature constraint.

[0015] A second aspect of this invention discloses a multi-machine cooperative decision-making control system for robot navigation, the system comprising: The acquisition module is used to determine at least one local machine from multiple machines, acquire local navigation information, and receive cooperative information from at least one other machine; The relationship module is used to determine the relative state relationship and path association between the local machine and any other machine based on the local navigation information and cooperation information; The conflict module is used to identify the communication interaction relationship between the local machine and other machines based on relative state relationships, path association relationships, and real-time control parameters to determine the conflict type in multi-machine scenarios. The decision module is used to define the multi-machine conflict intensity according to the conflict type in the multi-machine scenario, and to construct local scenario constraints by acquiring environmental obstacle information. Based on the multi-machine conflict intensity and local scenario constraints, it determines the local cooperative passage decision result in the current scenario. The control module is used to generate constraint information for local planning and motion control based on the collaborative passage decision results, so as to control the local machine and other machines to complete collaborative passage within the shared passage space.

[0016] A third aspect of this invention discloses another multi-machine cooperative decision-making control system for robot navigation, the system comprising: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute some or all of the steps in the multi-machine cooperative decision control method for robot navigation disclosed in the first aspect of the present invention.

[0017] The fourth aspect of the present invention discloses a computer storage medium storing computer instructions, which, when invoked, are used to execute some or all of the steps in the multi-machine cooperative decision control method for robot navigation disclosed in the first aspect of the present invention.

[0018] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: This invention, by acquiring local navigation information and other-machine collaboration information and determining the relative state and path relationships between the local and other machines, changes the passive obstacle avoidance mode that only treats non-local machines as obstacles. It enables the local machine to actively understand and utilize the movement intentions and path planning of other machines, providing a sufficient information foundation for subsequent collaborative decision-making. This effectively avoids short-sighted decisions and globally ineffective maneuvers caused by only perceiving the current position, thus significantly improving the effectiveness of global decision-making. By identifying traffic interaction relationships based on relative state relationships, path relationships, and real-time control parameters, it determines the specific conflict types in multi-machine scenarios and further defines the multi-machine conflict intensity, providing differentiated solutions for complex multi-machine scenarios such as oncoming traffic, lane obstruction in the same direction, and intersection crossing. The decision-making basis is based on the classification and hierarchical processing of conflicts through refined and differentiated collaborative strategies. This eliminates the oscillations and traffic deadlocks caused by the homogenization of decision-making rules in multi-machine game. The intensity of multi-machine conflict is combined with environmental obstacle information to construct local scene constraints, and the collaborative passage decision results are determined based on this. Finally, the global collaborative decision is transformed into constraint information for local planning and motion control, eliminating the disconnect between multi-machine collaborative decision-making and single-machine motion control. This allows the upper-level multi-machine collaborative logic to be smoothly injected into the lower-level execution layer, ensuring that the machine strictly adheres to the collaborative agreement when avoiding obstacles and generating trajectories locally. This achieves the connection from macro-coordination to micro-execution, thereby realizing safe and efficient orderly passage of multi-machines in the shared space. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating a multi-machine collaborative decision-making and control method for robot navigation disclosed in an embodiment of the present invention.

[0021] Figure 2 This is a schematic diagram of the structure of a multi-machine collaborative decision control system for robot navigation disclosed in an embodiment of the present invention.

[0022] Figure 3 This is a schematic diagram of another multi-machine collaborative decision control system for robot navigation disclosed in an embodiment of the present invention; Figure 4 This is a schematic diagram of pose calculation and path relationship determination in a multi-machine cooperative decision control method for robot navigation disclosed in an embodiment of the present invention; Figure 5This is a schematic diagram of the generation of contour points and separator line constraints in a multi-machine collaborative decision control method for robot navigation disclosed in an embodiment of the present invention. Detailed Implementation

[0023] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0025] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0026] This invention discloses a multi-machine cooperative decision-making control method and system for robot navigation. By acquiring the robot's own navigation information and the cooperative information of other machines, and determining the relative state relationship and path association relationship between the robot and other machines, it changes the passive obstacle avoidance mode that only treats non-machines as obstacles. This allows the robot to actively understand and utilize the movement intentions and path planning of other machines, providing a sufficient information foundation for subsequent cooperative decision-making. It effectively avoids short-sighted decisions and globally ineffective maneuvers caused by only perceiving the current position, thus significantly improving the effectiveness of global decision-making. By identifying traffic interaction relationships based on relative state relationships, path association relationships, and real-time control parameters, it determines the specific conflict types in multi-machine scenarios, and further defines the multi-machine conflict intensity, providing guidance for oncoming traffic, lane obstruction in the same direction, and intersection crossing disputes. This approach provides differentiated decision-making support for complex multi-machine scenarios. Through refined and differentiated collaborative strategies, it classifies and grades conflicts, eliminating the game-theoretic turbulence and traffic deadlock caused by homogeneous decision rules. It combines multi-machine conflict intensity with environmental obstacle information to construct local scenario constraints, and determines collaborative passage decisions based on these constraints. Ultimately, it transforms global collaborative decisions into constraint information for local planning and motion control, eliminating the disconnect between multi-machine collaborative decision-making and single-machine motion control. This allows the upper-level multi-machine collaborative logic to be smoothly injected into the lower-level execution layer, ensuring that the machine strictly adheres to collaborative agreements during local obstacle avoidance and trajectory generation. This achieves seamless integration from macro-coordination to micro-execution, enabling safe and efficient orderly passage of multiple machines within a shared space. The following sections provide detailed explanations.

[0027] Example 1 Please see Figure 1 , Figure 4 and Figure 5 , Figure 1 This is a flowchart illustrating a multi-machine cooperative decision-making and control method for robot navigation disclosed in an embodiment of the present invention. Figure 1 The described multi-machine collaborative decision-making and control method for robot navigation can be applied to data processing systems / data processing equipment / data processing servers (wherein, the server includes local processing servers or cloud processing servers). For example... Figure 1 As shown, the multi-machine cooperative decision-making and control method for robot navigation can include the following operations: 101. Determine at least one local machine from among multiple machines, obtain local navigation information, and receive cooperative information from at least one other machine.

[0028] Optionally, the local navigation information includes the local machine's ID, timestamp, map name, local path, target point, robot model identifier, task status, multi-machine status, scene information, and real-time control parameters.

[0029] Optionally, collaborative information includes the other machine's ID, timestamp, map name, local path, target point, robot model identifier, task status, multi-machine status, scene information, and real-time control parameters.

[0030] Optionally, the cooperation information is sent by any of the other machines via broadcast and received by the local machine.

[0031] Optionally, the broadcast type of the other machine broadcasting cooperative information may include, but is not limited to, near-field communication broadcast and network broadcast.

[0032] Optionally, the multi-machine state can be normal state, locked state, escape state, and rotating state.

[0033] Optionally, a multi-machine system includes at least one local machine and one other machine, thus multi-machine scenarios may involve double-click decision-making or multi-machine decision-making.

[0034] 102. Based on the local navigation information and cooperation information, determine the relative state relationship and path association relationship between the local machine and any other machine.

[0035] 103. Based on relative state relationships, path association relationships, and real-time control parameters, identify the communication and interaction relationships between the local machine and other machines to determine the conflict type in multi-machine scenarios.

[0036] 104. Define the multi-machine conflict intensity according to the conflict type in the multi-machine scenario, and construct the local scenario constraints by obtaining environmental obstacle information. Determine the local cooperative passage decision result in the current scenario based on the multi-machine conflict intensity and the local scenario constraints.

[0037] 105. Generate constraint information for local planning and motion control based on the collaborative passage decision results, so as to control the local machine and other machines to complete collaborative passage in the shared passage space.

[0038] As can be seen, the above-described embodiments of the invention, by acquiring local navigation information and other-machine collaborative information and determining the relative state relationship and path association relationship between the local machine and other machines, change the passive obstacle avoidance mode that only treats non-local machines as obstacles. This enables the local machine to actively understand and utilize the movement intentions and path planning of other machines, providing a sufficient information foundation for subsequent collaborative decision-making. It effectively avoids short-sighted decisions and globally ineffective maneuvers caused by only perceiving the current position, thereby significantly improving the effectiveness of global decision-making. By identifying traffic interaction relationships based on relative state relationships, path association relationships, and real-time control parameters, it determines the specific conflict type in multi-machine scenarios and further defines the multi-machine conflict intensity, providing solutions for complex multi-machine scenarios such as oncoming traffic, lane obstruction in the same direction, and intersection crossing. Differentiated decision-making criteria, through refined and differentiated collaborative strategies for conflict classification and hierarchical processing, eliminate multi-machine game oscillations and traffic deadlocks caused by homogeneous decision-making rules. By combining multi-machine conflict intensity with environmental obstacle information to construct local scene constraints, and based on this, determining the collaborative passage decision results, the global collaborative decision is ultimately transformed into constraint information for local planning and motion control. This eliminates the disconnect between multi-machine collaborative decision-making and single-machine motion control, allowing the upper-level multi-machine collaborative logic to be smoothly injected into the lower-level execution layer. This ensures that the machine strictly adheres to the collaborative agreement when avoiding obstacles and generating trajectories locally, achieving seamless connection from macro-coordination to micro-execution, thereby enabling safe and efficient orderly passage of multiple machines in a shared space.

[0039] like Figure 4 As shown, in an optional embodiment, the step of determining the relative state relationship and path association between the local machine and any other machine based on the local navigation information and cooperation information includes: Calculate the longitudinal relative distance, lateral relative distance, and heading difference between the machine and any other machine based on the machine's attitude and the attitudes of other machines, and obtain the relative state relationship between the machine and all other machines. The target vector from the local position to the target point is calculated based on the local target point and local pose in the local navigation information. The relative vector from the local machine to the other machine is calculated based on the local target point and the other machine's position. The path judgment value of the local machine is obtained by performing a dot product between the target vector and the relative vector. The hybrid product is calculated based on the local pose, the other machine pose, and the target point, as well as the Euclidean distance between the local machine and the target point. The path planning value is obtained by the ratio of the hybrid product to the Euclidean distance. The path judgment value determines the front-to-back relationship of other machines relative to the machine in the forward direction, and the path planning value determines the lateral deviation of other machines from the machine's path. The path association relationship between the machine and any other machine is determined by the front-to-back relationship and the lateral deviation.

[0040] As an example, relative state relationships are expressed through the local pose. and its position and posture The calculation process is as follows: ; ; ; ; In the formula, This indicates the machine's position coordinates in the global coordinate system. Indicates the heading angle of the aircraft. This indicates the position coordinates of the machine in the global coordinate system. This indicates the heading angle of the aircraft. This indicates the position difference between this machine and other machines along the global x-axis. This indicates the position difference between this machine and other machines along the global y-axis. This indicates the longitudinal relative distance between this machine and the machine itself. This indicates the lateral relative distance between the other machine and the machine.

[0041] As an example, path associations are achieved through local poses. Other machine position and posture and local target point The calculation process is as follows:

[0042] ; ; ; ; ; In the formula, This represents the position coordinates of the target point in the global coordinate system. This represents the target vector from the local location to the target point. This represents a relative vector pointing from the local machine to another machine. This represents the path determination value used in dot product calculations to determine the relationship between the local machine and other machines. This represents the mixed product used to calculate the lateral deviation between this machine and other machines. This represents the Euclidean distance between the machine and the target point. This represents the path planning value used to determine the degree of lateral deviation of other machines from the path of the local machine.

[0043] Optionally, when the path judgment value is greater than 0, it can be determined that the other machine is located in the area in front of the machine in the direction of travel. Conversely, when the path judgment value is less than or equal to 0, it can be determined that the other machine is located in the area behind the machine in the direction of travel.

[0044] Optionally, when the path planning value is less than the path occupancy threshold, it can be determined that the lateral deviation of other machines from the local machine's path is small and that other machines are close to the local machine's path planning channel. Conversely, when the path planning value is greater than the path occupancy threshold, it is determined that other machines do not interfere with the local machine's path. The path occupancy threshold can be set by calculation or prior experience.

[0045] As an optional embodiment, the steps described above, including identifying the communication interaction relationship between the local machine and other machines based on relative state relationships, path association relationships, and real-time control parameters to determine the conflict type in a multi-machine scenario, include: The geometric parameters of the robot and other robots are determined based on the robot model identifiers, and the robot geometric model is constructed. The geometric parameters include the robot's forward length, backward length, left width, and right width. The minimum safe distance is calculated based on the geometric parameters of the machine and other machines and the preset collision safety margin. The collision risk between the machine and other machines is identified by the relationship between the minimum safe distance and the relative state. Set interaction thresholds for corresponding conflict types for communication interaction relationships, and calculate the difference in direction angle between the other machine and the local machine. The conflict types include perpendicular intersection, same-direction lane occupation, and oncoming traffic meeting. When the absolute difference between the direction angle and the vertical angle is less than the first interaction threshold, the conflict type corresponding to the passage interaction relationship between the local machine and other machines is determined to be vertical intersection. When the difference in direction angles is less than the second interaction threshold and the longitudinal relative distance is less than the minimum safe distance, the conflict type corresponding to the passage interaction relationship between the local machine and other machines is determined as same-direction lane occupation. When the absolute difference between the direction angle and the horizontal angle is less than the third interaction threshold, and the path planning value is less than the minimum safe distance, the conflict type corresponding to the passage interaction relationship between the local machine and other machines is determined as oncoming traffic.

[0046] As an example, the minimum safe distance is calculated using the machine's geometric parameters, the other machine's geometric parameters, and a preset collision safety margin. The calculation process is as follows:

[0047] ; In the formula, Indicates the minimum safe distance in the longitudinal direction. Indicates the minimum safe distance in the lateral direction. This represents the forward length in the machine's geometric parameters. Indicates the forward length of the machine. This indicates the longitudinal (along the aircraft's heading) collision safety margin. This refers to the left width in the machine's geometric parameters. This refers to the left-side width in the machine's geometric parameters. This indicates the collision safety margin in the lateral direction (perpendicular to the aircraft's heading).

[0048] Furthermore, when and Only when it is established can it be identified and judged that there is a safety collision risk when another machine enters the local area; otherwise, there is no safety collision risk. It is understandable that the preset collision safety margin can be set through experimental calculation or prior experience.

[0049] Specifically, the difference in direction angles is expressed as: ,in This indicates the path direction angle of the other machine (in the global coordinate system, it usually refers to the tangent direction of its planned path or the current driving direction). Indicates the path direction angle of this machine. It represents the angle difference between the paths of both parties and is used to determine the relative driving direction relationship.

[0050] Specifically, in the case of conflict types involving same-direction lane occupation, the relationship between the local machine and other machines can be further determined by the path judgment value, thereby determining whether the conflict type belongs to the case of forward lane occupation or backward insertion in same-direction lane occupation.

[0051] As an example, the criteria for determining the type of conflict are as follows: when At that time, it can be determined that the paths of this machine and other machines intersect approximately perpendicularly, where Indicates the first interaction threshold; when and It can be determined that both parties are in a lane-occupying relationship in the same direction, among which Indicates the second interaction threshold. Indicates the maximum safe distance; when and At that time, it can be determined that the two vehicles are in a relationship of passing each other in opposite directions. Indicates the third interaction threshold; It is understandable that the interaction thresholds set according to different types of traffic interaction relationships can be obtained through experimental calculations or prior experience. For example, the first interaction threshold is used to determine whether the path angle is close to 90°.

[0052] like Figure 5As shown, in an optional embodiment, the steps above, including defining the multi-machine conflict intensity based on the conflict type in the multi-machine scenario, constructing local scenario constraints by obtaining environmental obstacle information, and determining the local-based cooperative passage decision result in the current scenario based on the multi-machine conflict intensity and the local scenario constraints, include: The multi-machine conflict intensity is defined by a weighted sum of the number of robots corresponding to all conflict types and the weight of each conflict type. Obtain the obstacle point cloud in the local obstacle environment, and count and count the point cloud of each obstacle region based on the preset saturation upper limit to obtain the local scene constraints. The obstacle regions in the obstacle environment include the forward region, the left region, and the right region. The maximum allowable linear speed of the robot is determined based on the intensity of multi-machine conflict and the preset deadlock threshold and speed limit threshold. The robot's search replanning flag is determined based on the constraints of the local scene and the preset constraint threshold. The robot's multi-machine state factors are determined based on its own navigation information and collaborative information, and the task state factors of the robot and other robots are determined based on its own navigation information and collaborative information. For any robot, the decision priority is calculated using a priority function based on the maximum permissible linear velocity, the search replanning flag, the multi-machine state factor, and the task state factor. Based on decision priority, output passage decisions for robots in multi-robot scenarios, as well as collaborative passage decisions under all passage decisions of the robot itself.

[0053] As an example, the intensity of multi-machine conflict can be expressed as: ; In the formula, Indicates the intensity of multi-machine conflict. This indicates the number of robots involved in vertical intersection conflicts. This indicates the number of robots occupying the lane in the forward direction within the same-direction lane-occupying conflict type. This indicates the number of robots inserted backwards in a same-direction lane-occupying conflict type. This indicates the number of robots involved in a collision between oncoming vehicles. , , , This indicates the weight corresponding to the above conflict types.

[0054] Furthermore, when the multi-machine conflict intensity R is not less than the deadlock threshold When this occurs, it can be determined that there is a risk of deadlock among multiple machines in the current scenario. Typically, the cooperative passage decision of a single machine is to enter a stopping or waiting state. When the conflict intensity R between multiple machines is between the speed limit threshold and the deadlock threshold, i.e. If it is determined that there is a moderate risk of passage for multiple machines in the current scenario, the machine's collaborative passage decision will enter a speed-limited passage state.

[0055] As an example, the native scenario constraint can be represented as: ; In the formula, E represents the decision value of the local scene constraint. This represents the point cloud count statistics within the left-hand region. This represents the point cloud count statistics for the forward region. This represents the point cloud count statistics for the right-hand region. Table i shows the weight coefficients corresponding to the left region. This represents the weight coefficient corresponding to the forward region. This represents the weighting coefficient corresponding to the right-hand region.

[0056] Specifically, the scene constraints are based on the robot's body coordinate system. The effective obstacle point cloud is counted and statistically analyzed according to the left, front, and right regions, and the count values ​​are marked as follows: , and When an obstacle point falls into a certain area, the count value is incremented by one. The count value of each area adopts a saturation counting method, and the count value of each area has a preset saturation upper limit. When the count value of a certain area reaches the preset upper limit, it will no longer be accumulated to reduce the excessive influence of dense point clouds or noise points on the statistical results. When the count values ​​of the left area and the right area both exceed the preset density threshold, the judgment result of the restricted narrow area is output; otherwise, the judgment result of the wide area is output.

[0057] Specifically, when the determination value E of the local scene constraint is greater than the preset constraint threshold... If the current area where the local machine is located is determined to be a restricted area, and the determination value E of the local scene constraint is less than or equal to the constraint threshold, then... If the current area is unrestricted, the system will determine whether the local machine is in a restricted area. If the local machine is in a restricted area, the system will suppress search replanning for the local machine's passage decision. If the local machine is in an unrestricted area, the system will allow search replanning for the local machine's passage decision.

[0058] Specifically, in scenarios with only two robots (the local robot and the other robot), the dual-robot decision-making process is executed as follows: First, it is determined whether the other robot is rotating in place. If the robot in front is rotating in place, the local robot stops and waits. Then, it is determined whether the two robots are approximately moving in the same direction. If they are moving in the same direction, the robot behind follows. Second, when the robot is in a narrow passage or through a doorway in a restricted area, it has priority to pass. Third, it is determined whether the two robots are approximately perpendicular. If they are perpendicular, the smaller robot, determined by the robot's geometric model, waits, while the larger robot has priority to pass. Finally, it is determined whether the two robots are in a meeting relationship. If they are in a meeting relationship, both robots reduce their speed to a preset low speed value, and based on the robot's body coordinate system, a virtual obstacle separator is inserted between the two robots for detour.

[0059] Specifically, in scenarios involving multiple robots, a multi-robot decision-making process is executed. For each robot, its relative position, path direction difference, and minimum path distance with the host robot are calculated, and the number of robots involved in vertical intersection conflicts is counted. Number of robots blocking the road in front Number of robots inserted in reverse Number of robots meeting oncoming traffic Using the conflict intensity index, when Exceeding the deadlock threshold When this occurs, the machine enters a lock-up risk control state, reducing the maximum linear speed to zero and marking the multi-machine state as locked. Between the speed limit threshold With the deadlock threshold When there is a single cross traffic conflict and the machine's priority is lower than that of another machine, the machine enters a waiting state.

[0060] As an example, the process of calculating decision priorities using priority functions can be represented as follows: ; In the formula, Q represents the decision priority. This indicates the maximum permissible linear velocity. Represents the task status factor. Represents the multi-machine state factor. This indicates a search replanning flag (which also includes access factors for gate states or special areas).

[0061] Specifically, priority is used to characterize whether the local machine should wait or have priority in passing under the current conflict relationship, and the search replanning flag is used to control whether the planning module of the upper-level local path allows the search for other alternative paths.

[0062] It is understandable that in a multi-robot scenario, any robot can participate in the traffic decision-making and planning as the local unit. In this embodiment, all robots have planned local paths, which can be obtained through local navigation information and collaborative information. However, in actual multi-robot scenarios, the planned local paths are not suitable for traffic relationships with multiple conflict types. In particular, in multi-robot scenarios with more than two robots, further decision-making is required to ensure traffic efficiency.

[0063] Specifically, a decision priority is calculated for each robot. Therefore, for each robot that can act as the local robot, the local decision priority is calculated and compared with the decision priorities of other robots. When the decision priority of another robot is greater than that of the local robot, the local robot enters a waiting or yielding state. Otherwise, the local robot maintains priority passage. For each robot acting as the local robot, based on its local navigation information, local path, and actual control parameters, the maximum allowable linear speed, search replanning flag, whether to stop and wait, priority flag, and collaborative decision-making results between the local robot and multiple robots are provided.

[0064] As an optional embodiment, the step described above, generating constraint information for local planning and motion control based on the cooperative passage decision results to control the local machine and other machines to complete cooperative passage within the shared passage space, includes: Based on the geometric model of the other machine, generate contour vertices for local planning for the other machine in the case of collision and path conflict; The contour vertices are transformed to the global coordinate system on the other machine's pose and sampled along the contour boundary to form the other machine's global contour point set. The other machine's global contour point set is regarded as the dynamic obstacle boundary input to the local planning map as obstacle constraint information. The target local planning boundary line point set is generated based on the local pose and the pose of other local machines. The local boundary line point set serves as the boundary line constraint information in the local planning map. The target local machine is the key local machine in the vehicle meeting conflict and path conflict. Based on the real-time control parameters of the machine, the motion working domain under the longitudinal speed limit and curvature coordination control constraints is determined. Within the selected range of the motion working domain, the longitudinal speed limit based on the curvature constraint information of the motion trajectory is used to maintain curvature within the maximum allowable linear velocity. Based on obstacle constraint information, separator line constraint information, and curvature constraint information, continuous control is performed on the cooperative passage of the local machine and other machines within the shared passage space.

[0065] Understandably, the global contour point set of the other machine is generated based on its geometric model and pose. For the local machine, the other machine can be regarded as a dynamic obstacle for obstacle avoidance. Therefore, the global contour point set is regarded as an obstacle point set and input into the local planning map to be displayed as an obstacle. This allows the local machine to make real-time judgments during local planning. At the same time, based on the other machine being regarded as a dynamic obstacle, virtual dividing lines are further generated for the local planner to prevent the local planner from generating motion trajectories that cross the other machine in high-risk areas (restricted areas constrained by the local scene).

[0066] As an optional embodiment, the step of generating the set of separator points for the target local planning based on the local pose and the other machine pose in the above steps includes: Calculate the path orientation angle of the target machine based on the continuous poses of the target machine and other machines, and transform the poses of other machines into the path coordinate system of the target machine based on the path orientation angle to obtain the forward distance and lateral offset distance of other machines relative to the target machine. The effective conditions for generating the separator line range are preset, and the safety conditions for the separator line width are determined based on the geometric models of the local machine and other machines. Under the effective conditions and safety conditions, the slope of the separator line point set is calculated based on the forward distance and the lateral offset distance. The lateral coordinate constraints of the dividing line point set are determined based on the classification of the lateral position and lateral offset distance of other machines relative to the local machine. Generate the set of dividing line points for the target machine under local planning based on slope and lateral coordinate constraints.

[0067] As an example, the process of generating the path direction angle is as follows: Get local pose and its position and posture In the example, 10 points are collected from the local path points to obtain the continuous pose of the local machine, and the local path orientation angle is calculated: ; in, This is the starting point of the current path on this machine. This is the 10th path point on the local machine path. The direction angle of the starting path point. This is the direction angle of the current path point.

[0068] As an example, the process for generating the forward distance and lateral offset distance is as follows: Transform its position and pose to the local path coordinate system: ; In the formula, This indicates the forward distance of the other machine relative to the first machine along the path. This indicates the lateral offset distance of the other machine relative to the path direction of the local machine.

[0069] Optionally, the valid conditions for pre-setting the range of separator lines can be: The unit for valid conditions is meters.

[0070] Optionally, the safety conditions for the separator line width can be... It is easy to know that among them Half the width of the machine. It is half the width of the machine.

[0071] As an example, the process of generating the slope of the separator point set is represented as follows: ; ; ; Specifically, the classification of the lateral position and lateral offset distance of the other machine relative to the machine is as follows: When the other machine is located to the left of the machine, the lateral offset distance... The horizontal coordinate of the separator line is The horizontal coordinate constraint is .

[0072] Furthermore, when the other machine is located to the right of this machine, The horizontal coordinate of the separator line is The horizontal coordinate constraint is Therefore, based on the slope and lateral coordinate constraints, the set of dividing line points of the target machine under local programming is obtained. Similar to obstacle avoidance constraint information, it is used by the local planner.

[0073] As an optional embodiment, the above steps, including determining the motion working domain under longitudinal speed limit and curvature coordination control constraints based on the real-time control parameters of the machine, and performing longitudinal speed limit while maintaining curvature within the maximum allowable linear velocity based on the curvature constraint information of the motion trajectory within the selected range of the motion working domain, include: By comparing the preset minimum linear velocity threshold and rotational angular velocity threshold with the linear velocity and angular velocity in the real-time control parameters, the motion working domain corresponding to the real-time control parameters of the machine is determined. The curvature constraint of the motion trajectory is defined according to the motion working domain, where the curvature is the rate of change of heading per unit path length when the machine travels along the current motion trajectory; The longitudinal linear velocity is limited and adjusted based on the maximum permissible linear velocity in the collaborative passage decision results, and the angular velocity in the actual control parameters is corrected based on the longitudinal linear velocity after the limit adjustment according to the curvature constraint.

[0074] It should be noted that the longitudinal speed limit based on curvature preservation in this embodiment is used to ensure the continuity, smoothness and predictability of the robot's motion trajectory during multi-machine cooperative passage. The motion working domain corresponds to the selection range for limiting the speed of real-time control parameters (speed control) when the robot is under non-lateral movement constraint conditions or non-stationary rotation constraint conditions.

[0075] Specifically, the non-lateral translation constraint means that the robot is not currently performing lateral translational motion, that is, the robot's lateral linear velocity is 0. The non-stationary rotation constraint means that the robot is not currently in a state of stationary turning or approximately stationary rotation.

[0076] Specifically, in this embodiment, the input of the local real-time control parameters is represented as follows: ,in This indicates the robot's longitudinal linear velocity. This indicates the robot's lateral linear velocity. This represents the robot's angular velocity.

[0077] As an example, longitudinal speed limiting and curvature coordination control acts on the following motion domain: ; In the formula, This represents the minimum lateral linear velocity threshold. This represents the minimum longitudinal linear velocity threshold. This represents the threshold value for angular velocity during rotation.

[0078] Furthermore, under the conditions of the aforementioned motion working domain, the curvature constraint can be defined as:

[0079] in, This represents the rate of change of heading per unit path length as the machine travels along its current trajectory. This indicates the longitudinal linear velocity of the machine. This indicates the machine's angular velocity.

[0080] It is understandable that the maximum permissible linear velocity has already been determined in the current scenario based on the multi-machine conflict intensity and the local scene constraints to determine the cooperative passage decision results based on the local machine. Therefore, the longitudinal linear velocity can be limited and adjusted according to the maximum permissible linear velocity.

[0081] As an example, the limiting adjustment of the longitudinal linear velocity can be expressed as: ; In the formula, This indicates the longitudinal linear velocity after the speed limit is applied. The sign function is used to preserve the direction of the original velocity. This represents the minimum value.

[0082] Furthermore, to ensure that the robot's trajectory geometry remains essentially consistent before and after the speed limit, the curvature can be maintained. The angular velocity remains unchanged, and is adjusted accordingly. The adjustment process can be expressed as follows:

[0083] In the formula, For the corrected angular velocity, it should be noted that, based on the speed limit while maintaining curvature, when the robot is not performing lateral movement and is not rotating in place, this embodiment does not independently trim the longitudinal velocity and angular velocity separately. Instead, it scales the longitudinal linear velocity and angular velocity in a linked manner based on the original trajectory curvature. This avoids over-steering, trajectory distortion, or local oscillations caused by only reducing the linear velocity while keeping the original angular velocity unchanged, thereby improving the trajectory continuity during multi-robot cooperative passage.

[0084] As a further example, when the collaborative decision-making result indicates that the current level of conflict is high, the passage is significantly restricted, or the local robot's priority is lower than that of the target robot, the maximum longitudinal linear velocity can be further reduced to zero, i.e.: At this point, the real-time control parameters are directly taken as: This allows the robot itself or other robots to enter a stopping and waiting state, a yielding state, or a locked-out avoidance state.

[0085] Specifically, in this embodiment, the maximum permissible speed can be dynamically given by the multi-machine collaborative decision-making results, where the variable indicators are decision priority Q, local scenario constraint E, and multi-machine conflict intensity R, and therefore can be expressed by a functional relationship as follows; ; Generally, when the intensity of multi-machine conflict or the value of local scene constraints increases, the maximum allowable linear velocity decreases. When the local machine's priority is lower than that of other machines (i.e., the value of Q decreases), the maximum allowable linear velocity will also be further reduced so that the local machine can prioritize performing deceleration or stopping and waiting actions.

[0086] Furthermore, it can be seen that under conditions of non-lateral movement and non-rotational constraints, by limiting the longitudinal linear velocity and synchronously adjusting the angular velocity according to the curvature preservation principle, the robot can smoothly move along its original trajectory even as its speed decreases. This control method is particularly suitable for situations such as multiple robots meeting, avoiding obstacles, slowing down to block obstacles ahead, and passing in narrow passages, which is beneficial for achieving continuous collaboration and smooth passage among multiple robots.

[0087] In a specific implementation scheme, a technical solution based on an embodiment of the present invention is implemented. In summary, the technical solutions disclosed in the embodiments of the present invention have the following advantages: 1. Accurate Conflict Identification: By simultaneously incorporating robot broadcast paths, real-time status, environmental obstacle point clouds, and map information into multi-machine decision-making, the accuracy of conflict identification is improved.

[0088] 2. Complex Scene Modeling: Modeling of dual-machine or multi-machine interactions can cover complex scenarios such as oncoming traffic, following in the same direction, perpendicular intersections, lane blocking ahead, and multi-vehicle congestion.

[0089] 3. Clear passage boundaries: Using robot geometric models to establish more accurate depictions of real passage boundaries.

[0090] 4. Decision consistency: By generating contour point sets and separator line constraints, the decision results of the upper-level multi-machine system can be directly applied to the local planner, improving the consistency of decision execution.

[0091] 5. Reduce oscillations: By linking search replanning markers with environmentally restricted area determination, ineffective replanning and path oscillations in restricted areas can be reduced.

[0092] 6. Flexible collaboration: By using waiting states and priority strategies, the probability of multi-machine stalemate and deadlock can be reduced.

[0093] 7. Smooth trajectory: Curvature-maintaining speed limiting control can improve the smoothness of the trajectory and control stability during deceleration.

[0094] 8. Formulated Indicators: By using formulated conflict indicators, scenario constraints, and priority functions, the interpretability, reproducibility, and feasibility of the solution are improved.

[0095] Example 2 Please see Figure 2 , Figure 2 This is a schematic diagram of a multi-machine collaborative decision-making control system for robot navigation disclosed in an embodiment of the present invention. Figure 2 The described multi-machine collaborative decision-making control system for robot navigation can be applied to data processing systems / data processing equipment / data processing servers (wherein, the server includes local processing servers or cloud processing servers). For example... Figure 2 As shown, the multi-machine collaborative decision control system for robot navigation may include: The acquisition module 201 is used to determine at least one local machine from multiple machines, acquire local navigation information, and receive cooperative information from at least one other machine; The relationship module 202 is used to determine the relative state relationship and path association relationship between the local machine and any other machine based on the local navigation information and cooperation information; Conflict module 203 is used to identify the passage interaction relationship between the local machine and other machines based on relative state relationship, path association relationship and real-time control parameters to determine the conflict type in multi-machine scenario; The decision module 204 is used to define the multi-machine conflict intensity according to the conflict type in the multi-machine scenario, and to construct local scenario constraints by acquiring environmental obstacle information, and to determine the local cooperative passage decision result in the current scenario based on the multi-machine conflict intensity and the local scenario constraints. The control module 205 is used to generate constraint information for local planning and motion control based on the collaborative passage decision results, so as to control the local machine and other machines to complete collaborative passage in the shared passage space.

[0096] Example 3 Please see Figure 3 , Figure 3 This is another multi-machine collaborative decision control system for robot navigation disclosed in the embodiments of the present invention. Figure 3 The described multi-machine collaborative decision-making control system for robot navigation is applied in data processing systems / data processing equipment / data processing servers (wherein, the server includes local processing servers or cloud processing servers). For example... Figure 3 As shown, the multi-machine collaborative decision control system for robot navigation may include: Memory 301 storing executable program code; Processor 302 coupled to memory 301; The processor 302 calls the executable program code stored in the memory 301 to execute the steps of the multi-machine cooperative decision control method for robot navigation described in Embodiment 1.

[0097] Example 4 This invention discloses a computer read storage medium that stores a computer program for electronic data interchange, wherein the computer program causes a computer to execute the steps of the multi-machine cooperative decision control method for robot navigation described in Embodiment 1.

[0098] Example 5 This invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to perform the steps of the multi-machine cooperative decision control method for robot navigation described in Embodiment 1.

[0099] The foregoing has described specific embodiments of this specification; other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims may be performed in a different order than those shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily have to follow the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0100] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0101] For ease of description, the above devices are described in terms of function, divided into various units. Of course, in implementing this specification, the functions of each unit can be implemented in one or more software and / or hardware components.

[0102] Those skilled in the art will understand that the embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the embodiments of this specification can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0103] This specification is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this specification. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0104] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0105] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0106] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0107] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0108] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0109] It should also be noted that 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 limitation, 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.

[0110] This specification can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This specification can also be practiced in distributed computing environments, where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0111] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0112] Finally, it should be noted that the multi-machine collaborative decision-making control method and system for robot navigation disclosed in the embodiments of the present invention are merely preferred embodiments of the present invention and are only used to illustrate the technical solutions of the present invention, not to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi-machine cooperative decision-making and control method for robot navigation, characterized in that, The method includes: Identify at least one local machine from among multiple machines, obtain local navigation information, and receive cooperative information from at least one other machine; Based on local navigation information and cooperative information, determine the relative state relationship and path association between the local machine and any other machine; Based on relative state relationships, path association relationships, and real-time control parameters, the communication interaction relationships between the local machine and other machines are identified to determine the type of conflict in multi-machine scenarios. The multi-machine conflict intensity is defined based on the conflict type in the multi-machine scenario, and the local scenario constraints are constructed by acquiring environmental obstacle information. Based on the multi-machine conflict intensity and the local scenario constraints, the local cooperative passage decision result in the current scenario is determined. Based on the results of the collaborative passage decision, constraint information is generated for local planning and motion control to control the local machine and other machines to complete collaborative passage within the shared passage space.

2. The multi-machine cooperative decision-making and control method for robot navigation according to claim 1, characterized in that, The local navigation information includes the local machine's ID, timestamp, map name, local path, target point, robot model identifier, task status, multi-machine status, scene information, and real-time control parameters. The collaborative information includes the other machine's ID, timestamp, map name, local path, target point, robot model identifier, task status, multi-machine status, scene information, and real-time control parameters. The collaborative information is broadcast by any of the other machines and received by the local machine.

3. The multi-machine cooperative decision-making and control method for robot navigation according to claim 1, characterized in that, Based on local navigation information and cooperative information, determine the relative state relationship and path association between the local machine and any other machine, including: Calculate the longitudinal relative distance, lateral relative distance, and heading difference between the machine and any other machine based on the machine's attitude and the attitudes of other machines, and obtain the relative state relationship between the machine and all other machines. The target vector from the local position to the target point is calculated based on the local target point and local pose in the local navigation information. The relative vector from the local machine to the other machine is calculated based on the local target point and the other machine's position. The path judgment value of the local machine is obtained by performing a dot product between the target vector and the relative vector. The hybrid product is calculated based on the local pose, the other machine pose, and the target point, as well as the Euclidean distance between the local machine and the target point. The path planning value is obtained by the ratio of the hybrid product to the Euclidean distance. The path judgment value determines the front-to-back relationship of other machines relative to the machine in the forward direction, and the path planning value determines the lateral deviation of other machines from the machine's path. The path association relationship between the machine and any other machine is determined by the front-to-back relationship and the lateral deviation.

4. The multi-machine cooperative decision-making and control method for robot navigation according to claim 1, characterized in that, Based on relative state relationships, path association relationships, and real-time control parameters, the communication interaction relationships between the local machine and other machines are identified to determine the types of conflicts in multi-machine scenarios, including: The geometric parameters of the robot and other robots are determined based on the robot model identifiers, and the robot geometric model is constructed. The geometric parameters include the robot's forward length, backward length, left width, and right width. The minimum safe distance is calculated based on the geometric parameters of the machine and other machines and the preset collision safety margin. The collision risk between the machine and other machines is identified by the relationship between the minimum safe distance and the relative state. Set interaction thresholds for corresponding conflict types for communication interaction relationships, and calculate the difference in direction angle between the other machine and the local machine. The conflict types include perpendicular intersection, same-direction lane occupation, and oncoming traffic meeting. When the absolute difference between the direction angle and the vertical angle is less than the first interaction threshold, the conflict type corresponding to the passage interaction relationship between the local machine and other machines is determined to be vertical intersection. When the difference in direction angles is less than the second interaction threshold and the longitudinal relative distance is less than the minimum safe distance, the conflict type corresponding to the passage interaction relationship between the local machine and other machines is determined as same-direction lane occupation. When the absolute difference between the direction angle and the horizontal angle is less than the third interaction threshold, and the path planning value is less than the minimum safe distance, the conflict type corresponding to the passage interaction relationship between the local machine and other machines is determined as oncoming traffic.

5. The multi-machine cooperative decision-making and control method for robot navigation according to claim 1, characterized in that, The multi-machine conflict intensity is defined based on the conflict type in the multi-machine scenario, and local scenario constraints are constructed by acquiring environmental obstacle information. Based on the multi-machine conflict intensity and local scenario constraints, the local cooperative passage decision result in the current scenario is determined, including: The multi-machine conflict intensity is defined by a weighted sum of the number of robots corresponding to all conflict types and the weight of each conflict type. Obtain the obstacle point cloud in the local obstacle environment, and count and count the point cloud of each obstacle region based on the preset saturation upper limit to obtain the local scene constraints. The obstacle regions in the obstacle environment include the forward region, the left region, and the right region. The maximum allowable linear speed of the robot is determined based on the intensity of multi-machine conflict and the preset deadlock threshold and speed limit threshold. The robot's search replanning flag is determined based on the constraints of the local scene and the preset constraint threshold. The robot's multi-machine state factors are determined based on its own navigation information and collaborative information, and the task state factors of the robot and other robots are determined based on its own navigation information and collaborative information. For any robot, the decision priority is calculated using a priority function based on the maximum permissible linear velocity, the search replanning flag, the multi-machine state factor, and the task state factor. Based on decision priority, output passage decisions for robots in multi-robot scenarios, as well as collaborative passage decisions under all passage decisions of the robot itself.

6. The multi-machine cooperative decision-making and control method for robot navigation according to claim 1, characterized in that, Based on the cooperative passage decision results, constraint information is generated for local planning and motion control to control the local machine and other machines to complete cooperative passage within the shared passage space, including: Based on the geometric model of the other machine, generate contour vertices for local planning for the other machine in the case of collision and path conflict; The contour vertices are transformed to the global coordinate system on the other machine's pose and sampled along the contour boundary to form the other machine's global contour point set. The other machine's global contour point set is regarded as the dynamic obstacle boundary input to the local planning map as obstacle constraint information. The target local planning boundary line point set is generated based on the local pose and the pose of other local machines. The local boundary line point set serves as the boundary line constraint information in the local planning map. The target local machine is the key local machine in the vehicle meeting conflict and path conflict. Based on the real-time control parameters of the machine, the motion working domain under the longitudinal speed limit and curvature coordination control constraints is determined. Within the selected range of the motion working domain, the longitudinal speed limit based on the curvature constraint information of the motion trajectory is used to maintain curvature within the maximum allowable linear velocity. Based on obstacle constraint information, separator line constraint information, and curvature constraint information, continuous control is performed on the cooperative passage of the local machine and other machines within the shared passage space.

7. The multi-machine cooperative decision-making and control method for robot navigation according to claim 6, characterized in that, Generate a set of dividing line points for the target local planning based on the local pose and the poses of other machines, including: Calculate the path orientation angle of the target machine based on the continuous poses of the target machine and other machines, and transform the poses of other machines into the path coordinate system of the target machine based on the path orientation angle to obtain the forward distance and lateral offset distance of other machines relative to the target machine. The effective conditions for generating the separator line range are preset, and the safety conditions for the separator line width are determined based on the geometric models of the local machine and other machines. Under the effective conditions and safety conditions, the slope of the separator line point set is calculated based on the forward distance and the lateral offset distance. The lateral coordinate constraints of the dividing line point set are determined based on the classification of the lateral position and lateral offset distance of other machines relative to the local machine. Generate the set of dividing line points for the target machine under local planning based on slope and lateral coordinate constraints.

8. The multi-machine cooperative decision-making and control method for robot navigation according to claim 6, characterized in that, Based on the real-time control parameters of the machine, the motion operating domain under longitudinal speed limit and curvature coordination control constraints is determined. Within the selected range of the motion operating domain, longitudinal speed limit is performed to maintain curvature within the maximum allowable linear velocity, based on the curvature constraint information of the motion trajectory. This includes: By comparing the preset minimum linear velocity threshold and rotational angular velocity threshold with the linear velocity and angular velocity in the real-time control parameters, the motion working domain corresponding to the real-time control parameters of the machine is determined. The curvature constraint of the motion trajectory is defined according to the motion working domain, where the curvature is the rate of change of heading per unit path length when the machine travels along the current motion trajectory; The longitudinal linear velocity is limited and adjusted based on the maximum permissible linear velocity in the collaborative passage decision results, and the angular velocity in the actual control parameters is corrected based on the longitudinal linear velocity after the limit adjustment according to the curvature constraint.

9. A multi-machine cooperative decision-making control system for robot navigation, characterized in that, The system includes: The acquisition module is used to determine at least one local machine from multiple machines, acquire local navigation information, and receive cooperative information from at least one other machine; The relationship module is used to determine the relative state relationship and path association between the local machine and any other machine based on the local navigation information and cooperation information; The conflict module is used to identify the communication interaction relationship between the local machine and other machines based on relative state relationships, path association relationships, and real-time control parameters to determine the conflict type in multi-machine scenarios. The decision module is used to define the multi-machine conflict intensity according to the conflict type in the multi-machine scenario, and to construct local scenario constraints by acquiring environmental obstacle information. Based on the multi-machine conflict intensity and local scenario constraints, it determines the local cooperative passage decision result in the current scenario. The control module is used to generate constraint information for local planning and motion control based on the collaborative passage decision results, so as to control the local machine and other machines to complete collaborative passage within the shared passage space.

10. A multi-machine cooperative decision-making control system for robot navigation, characterized in that, The system includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the multi-machine cooperative decision control method for robot navigation as described in any one of claims 1-8.

Images