SLAM-BASED MULTIPLE MOBILE ROBOT TRAFFIC COORDINATION SYSTEM AND METHOD

Abstract

The invention specifically relates to a system and method for coordinating multiple mobile robots operating on two-dimensional navigation maps created using Simultaneous Localization and Mapping (SLAM), via a graphic-based and centralized decision-making mechanism.

Description

SLAM-BASED MULTIPLE MOBILE ROBOT TRAFFIC COORDINATION SYSTEM AND METHOD Technical Field to Which the Invention Relates The invention specifically enables simultaneous localization and mapping. two-dimensional navigation maps created using SLAM (Surface-to-Air Mapping) a graphical-based and centralized decision-making mechanism for multiple mobile robots It relates to a system and method for coordinating through. State of the Art In systems where multiple mobile robots perform tasks in the same physical workspace, each The robot plans its path to reach a target point, considering its surroundings. It needs to avoid obstacles and maintain a safe distance from other robots. In current applications, these processes involve local sensing and control running on the robot. algorithms, fleet management software, coordinate-based route planning methods, and This can be done in accordance with traffic rules relating to specific transit zones. In solutions where robots continuously plan their paths in coordinate space, a central... the control unit monitors the real-time coordinates and constantly changing routes of numerous robots The robot's ability to evaluate its geometries together can increase the computational load. As the number of shared corridors or junctions increases, the potential route interactions increase. The number of collision avoidance systems based solely on instantaneous physical proximity is also increasing. This approach is limited in its ability to anticipate future bottlenecks. It can remain. The two-dimensional maps obtained with SLAM serve as a completely free navigation space. In the applications where they are used, it is important to know in which areas the robots can move. how many robots can be in the area at the same time and which ones can be in a specific transit zone The issue is that the implementation of traffic policy is not modeled in a clear and distinct manner. This is possible. In this case, multiple robots may be used in the same narrow corridor or shared passage. The resulting traffic congestion, inefficient waiting, and gridlock at the designated point. This can lead to these situations. 1 In the known state of the art, area boundaries are determined by physical markers. Applying static capacity or transition rules for specific regions, graph nodes Time / access reservation and allocation to robots, and parking of conflicting robots. Approaches such as directing them to waiting areas may be found. These types of dependence on physical marker infrastructure in approaches, reservation data In case of workload or conflict related to its management, the robot directly A flow interruption may occur due to the cessation of flow. However, in order to manage multi-robot traffic in a predictive manner, the robots need to be... not only their current locations but also the route segments they plan to follow in the future centralized comparison; if a conflict is identified, robots are directly compared. Instead of stopping, alternative parallel routes are being developed to the extent that the physical lateral capacity of the region allows. the creation of transit routes; or, where this is not possible, predictable alternatives. And the need for a transition decision to be made through a deterministic priority mechanism continues. is doing. Brief Description and Objectives of the Invention The overall purpose of the invention is to create a two-dimensional navigation map based on SLAM. centrally controls the route interactions of multiple mobile robots sharing the workspace. monitoring and identifying potential collisions before robots physically approach each other. and a graph that intervenes in the identified overlap according to the geometric capacity of the region. The aim is to provide a traffic coordination system and method based on a solid foundation. One aim of the invention is to render the entire SLAM map as unsupervised free space. Instead of using, the robots' permitted movement areas are defined by the user. limiting navigation zones and establishing a capacity and usage policy for each zone. The goal is to provide structured traffic management across the system by assigning tasks. Another purpose of the invention is to enable navigation within user-defined zones. By creating a discrete graphical structure, robot positions and route plans are represented with graphical identifiers. to represent and centralize traffic assessment of continuous coordinate data The aim is to reduce the complexity it can create. Another purpose of the invention is to enable robots to follow user-defined routes along their planned paths. To centrally compare the graph IDs of k forward nodes and create a common graph. 2 If identification is found, potential route conflicts may occur through physical proximity. It means identifying it before it happens. Another aim of the invention is to directly intercept robots when a potential collision is identified. Instead of waiting, first consider the lateral free space capacity of the navigation region. to evaluate, if adequate capacity is available, the main navigation line laterally by shifting to create a new passage and thus allowing traffic to flow without interruption. or to maintain it with reduced need for posture. Another purpose of the invention is to address narrow or physical obstacles where line shifting is not possible. In single-robot navigation areas, the distance of the robots to the target or their advance a deterministic winner-loser decision using a defined priority rule The goal is to produce and thus reduce the likelihood of mutual deadlock. Another purpose of the invention is to enable the continuation, repetition produced by the central decision-maker. Robots can locally execute high-level behavioral commands such as planning, waiting, or parking. By transmitting this information to control and planning units, central traffic decisions are made at the robot level. The goal is to create a consistent flow of data and commands between physical movements. Detailed Description of the Invention The system described in the invention is a SLAM-based two-dimensional navigation map, manually defined. navigation zones, regional capacity and policy definition, graph creation module, graph node structure, graph ID conversion unit, central decision-making unit, collision, and congestion analysis module, prioritization mechanism, line shifting reconfiguration planning mechanism, higher-level behavioral command generation unit, and robot local control. and includes a planning unit. SLAM-based two-dimensional navigation maps are used by mobile robots or external sensors. created using data collected from the environment and locations in the study area It is a basic navigation map in which the surface is represented within a common coordinate system. The subject is the map, the positioning of robots, and the geometric representation of the work environment. It is used for representation. The invention is made with a specific SLAM algorithm. It is not limited, and the subject of the application is two inputs for traffic coordination. 3D map data is used. 3 Manually defined navigation zones are two-dimensional, SLAM-based, defined by the user. They are rectangular or polygonal geometric areas placed on a navigation map. These areas define the permitted navigation zones for robots and the map. This prevents the entire area from being used as an unrestricted space of free movement. The user can navigate through the main corridors, narrow passageways, loading areas, and cargo handling areas in the work environment. separate navigation zones from drop-off areas, intersections, or similar functional areas. It can be defined as such. Regional capacity and policy definition, simultaneously for each navigation region. the number of robots allowed and traffic rules regarding the use of the area This indicates that a navigation area can be configured with a single robot capacity. such as, allowing multiple robots to be present simultaneously when the geometric width permits. It can also be structured in a way that provides this capacity and policy information. Traffic rules to be applied by the central decision-making unit in case of conflict. It is used in determining behavior. The graphics generation module is used in the path planning of robots over navigation areas. This creates the discrete graph structure to be used. The graph structure consists of graph node structures. and consists of edges representing the connections between these nodes. The invention a graphing module in an application, user-defined rectangles or the midpoints of opposite edges of polygonal navigation regions It determines and establishes main navigation lines between these midpoints. Main navigational lines established in a preferred application of the invention, beforehand It is divided into successive segments according to a defined, fixed step size. Each segment or segment-associated graph node structure, in a hierarchical identity structure It is tagged with a unique graphic identifier. This hierarchical identifier includes the Regional ID, It includes Line ID and Segment Order components. Thus, the same study Different regions, lines, and segments within the area can be distinctly identified. The graphical identity conversion unit enables robots to navigate using SLAM-based two-dimensional technology. real-time coordinate information on the map, its location as a graph node, or It converts the graphic identity associated with the segment. Mobile robots also navigate the local path. A graph of current and future route plans obtained as a result of planning. It transmits the identity sequences to the central decision-making unit. Thus, the central system, 4 Not only where the robots are located, but also which graph they follow in the subsequent route steps. He can also track that he plans to use the pieces through discrete identity sequences. The collision and congestion analysis module is used to analyze data received by the central decision-making unit. It compares current and future graphic ID information. This comparison, It is not based solely on the principle of the robots' instantaneous coordinates converging; a forward-looking view of the graphical nodes that robots plan to track in the future An evaluation is underway. User for forward-looking evaluation in the preferred application of the invention. A forward-looking node number k is defined by the central decision-making unit. Each robot has k robots located further along the route, starting from its current graphic ID. It continuously scans for the identity of the graph node or segment. Collisions and congestion analysis module, for example, the route ID of the first robot to the next node k. comparing the set of route IDs for the next k nodes of the second robot with the set of route IDs for the next k nodes. and a potential route if there is at least one common graph identifier between the two sets. The overlap determines the process. This process involves the robots physically approaching each other. It can be done before. In applications involving multiple robots, this forward-looking comparison is central. route ID sequences of robots subject to traffic evaluation in the system is being executed. The user-defined value of k determines how many future graphs It is determined that the node will be included in the collision analysis, and the invention specifies a particular k It is not limited by its value. If a potential conflict is identified, the central decision-making unit will first investigate the line. The shifting mechanism evaluates the rescheduling mechanism. Line shifting. replanning mechanism, lateral of the navigation area where the overlap is located free space capacity depends on the geometric width of the area, the physical width of the robots, minimum safety distance required for safe passage and pre-programmed in the system It evaluates based on defined maximum scrolling limits. In a preferred application of the invention, two robots can be positioned side-by-side in the same navigation area. to determine if it is possible to cross safely The following capacity requirement is used: 2 × Wᵣ + 3 × Dₑ ≤ Wₚ Equation-1 Here, Wᵣ represents the physical width of the robot, Dₑ the minimum safety distance, and Wₚ This represents the width of the navigation polygon in which the collision is evaluated. inequality determines whether lateral capacity is suitable for implementing track shifting. It is a preferential control condition used to determine the amount of line shifting. The system keeps the scrolling within predefined maximum limits. If lateral capacity is sufficient, the track shift rescheduling mechanism and navigation It laterally offsets the main movement path used within the region and provides power to robots. It generates new transit route or replanning information. Thus, it avoids overlapping. This allows the robots to physically and safely share the same graphical area. Alternative movement routes can be created. The precise position of the line shifting operation. offset amount, lateral free space of the navigation region and predefined maximum The offset limits are determined by taking them into consideration, and the invention relates to a specific single offset. It does not depend on its value. New route created by implementing the route shift rescheduling mechanism information is transmitted to the relevant robots via a high-level behavioral command generation unit, robot locale. The information is transmitted to the control and planning unit. The robot's local control and planning unit, The newly transmitted line is physically equipped with its own local route planning and motion control functions. converting the movements into commands and directing the robot to advance along the new path. It provides. In narrow navigation areas where line shifting is physically impossible, or in regions where regional capacity and policy definition only allow for a single robot. The prioritization mechanism comes into play. The prioritization mechanism, Which of the robots involved in the collision will continue with its current plan and which one will... a deterministic winner-loser decision to determine whether to wait temporarily It produces. In a preferred application of the invention, the prioritization mechanism involves assigning a priority to each robot. It calculates the Euclidean distance between the current position and the target point. To the target The robot with the shorter distance is determined as the winner, and the current route... It receives authorization to continue with its plan. The other robot is classified as a loser. and a wait or park command by the higher-level behavioral command generation unit. 6 is being created. In an alternative application, the winning robot has a predefined priority. It can be determined according to the rule. The higher-level behavioral command generation unit evaluates the central decision-making unit. Depending on the outcome, continue, reschedule, wait, or park. It generates the appropriate command from the available commands. These commands enable the robots to exercise local control. It does not have to replace algorithms, but robot local control and planning. It provides its units with high-level behavioral limits or task instructions. The robot is local. Control and planning units will implement this high-level decision regarding local obstacle detection and speed. It is implemented using both control and motion control. Central decision-making unit, current graphic identities of robots, future roadmap Identity sequences, navigation area capacity and policy information, forward lookout node count and route shifting conditions are evaluated together in the traffic management process. Thus, decisions regarding robot traffic in the system are based on the local perception of a single robot. not being left behind, planned movements in the collaborative workspace are centralized at a higher level. It is being addressed under coordination. The invention works in an application with the following logical flow: This is being carried out. First, a SLAM-based two-dimensional navigation map is obtained. Navigation is being carried out and manually defined navigation zones are being created on the map. Regional capacity and policy definitions are being created for each region; graph generation module. Main navigation lines and graph node structures are created by this. The graphic ID conversion unit converts the robots' positions into graphic IDs. transforming and incorporating current and future graphics found in robots' route plans It transmits identity data to the central decision-making unit. Then, the collision and congestion analysis module determines k forward routes for each robot. It compares the identity sequences that encompass the node. Common graph identity If not found, the higher-level behavioral command generation unit will continue with the current plan. It produces a decision to act. If there is a common graphic identity, the central decision-maker The unit collision will occur in a navigation area with multiple robot passages physically. whether it is suitable as a line shift replanning mechanism is evaluating. 7 If the lateral capacity of the region is sufficient, the main navigation line is offset laterally. The replanned transit route is being communicated to the relevant robots, and traffic flow is being maintained. is provided. If the lateral capacity of the region is insufficient or the region has only one robot If there is capacity, a prioritization mechanism is activated, identifying winners and losers. The robot is being identified and won by the higher-level behavioral command generation unit. commands for the robot to continue, and for the losing robot to wait or park. It is manufactured. The robot's local control and planning units execute the commands and The system uses the same evaluation method with the current location and route information received from the robots. It repeats its cycle. In a concrete application of the invention, multiple units transport materials within a warehouse. The autonomous mobile robot performs tasks on the same SLAM-based two-dimensional navigation map. The warehouse features wide main corridors capable of handling multiple robots, while narrow passageways are designed for this purpose. They are defined as navigation zones with a single robot capacity. One robot for each zone. Capacity and usage policy are determined by the graph generation module. Discrete navigation lines are created along the corridors. The two robots' task paths share the same graph identity within the advancing k-node graph. If it contains a collision or blockage, the collision and blockage analysis module can detect potential collisions. This is determined by the fact that the overlap occurs in a wide corridor and the lateral capacity condition is met. If provided, the line shift replanning mechanism within the corridor the main line is shifted laterally and the robots need to constantly or halt to move each other. It creates a new transit route so that it can pass through in a reduced manner. The same overlap If it occurs in a narrow corridor, the prioritization mechanism targets the objective. Determining the winning robot based on distance or a defined priority rule; other robots They are temporarily held or directed to a parking position. Thanks to this technical structure, future route usage of robots will be determined using Graph ID arrays. can be compared via traffic intervention before physical proximity. It can be produced and in appropriate areas, it can respond to collisions with a direct halt. Lateral line shifting can be applied instead. User-defined region geometry, Discrete graph identity structure, forward-looking central overlap analysis, lateral capacity dependent. The combined operation of line shifting and deterministic prioritization mechanisms, It enables the predictable and centralized management of multiple robot traffic. 8 The invention is not limited to the warehouse application described above, but also includes in-factory production lines. material handling, semi-finished product transfer, inter-station logistics, loading and unloading between stations. drop-off operations using multiple mobile robots in shared passageways It can be used in other intralogistics applications. The scope of the invention is defined in the claims. This is determined by the specified technical specifications. 9

Claims

1. Multiple mobile devices operating on a SLAM-based two-dimensional navigation map. It is a system designed to ensure the robot's traffic coordination, and its feature is; − User-defined on a SLAM-based two-dimensional navigation map Manually defined navigation, described as a rectangle or polygon. regions; robot capacity and usage for those navigation regions regional capacity and policy definition that sets the rules; − Graphic node with discrete navigation lines on navigation regions The graphics creation module that creates the structures; − a graph that converts the real-time locations of robots into graphical identifiers Identity conversion unit; - Current graphic identities of robots and future route graphic identities. receiving the sequences and the route graph ID sequences in question by the user Comparing k defined forward graph nodes to find a common graph identity. collision and blockage that identify potential collisions if present Central decision-making unit containing an analysis module; − When the potential overlap in question is identified, the lateral aspect of the navigation area Assessing free space capacity to ensure lateral capacity is sufficient. offsetting the main navigation line laterally and creating a new passage for robots The line shifting replanning mechanism that forms the line; - where line shifting is not possible or the navigation area is a single robot If it has the capacity, there is a deterministic winner among the robots. The prioritization mechanism that produces the losing decision; - and proceeding or replanning according to the decision of the central decision-making unit, The robot must have local control and at least one of the waiting or parking commands. High-level behavioral command generation unit that transmits to the planning unit. It includes.

2. Multiple mobile devices operating on a SLAM-based two-dimensional navigation map. This is a method for coordinating robot traffic, and its feature is; − User-defined on a SLAM-based two-dimensional navigation map Definition of rectangular or polygonal navigation zones and region through capacity and policy definition for these navigation areas Determining the robot's capacity and usage rules. − Discrete navigation zones by the graph generation module Creating graphic node structures with navigation lines, − Real-time robots by the graphic ID conversion unit converting their locations into graphic identities, − by the central decision-making unit with the robots' existing graphic IDs Obtaining future route chart identity sequences, - Conflict and bottleneck analysis module located in the central decision-making unit. by, on the route from each robot's current graphic ID. graph belonging to k user-defined forward graph nodes comparing their identities in terms of different robots and finding at least one common one Identifying potential conflicts if graphic identification is found. - Line shifting and rescheduling if a potential overlap is identified. lateral free space of the relevant navigation region by the mechanism assessment of capacity and whether the said lateral capacity is sufficient If this happens, the main navigation line is laterally offset so that the robots... the creation of a new transit route, − where line shifting is not possible or the relevant navigation area is unique by the prioritization mechanism if the robot has the capacity a deterministic winner-loser between the robots involved in the collision making the decision, − Central decision-maker by the upper-level behavior command generation unit Continue, reschedule, wait, or park, depending on the unit's decision. At least one of the commands must be sent to the robot's local control and planning unit. transmission and − the command transmitted by the robot's local control and planning unit implementation of the robot through local path planning and motion control It includes the steps of the process.

3. A system that complies with Claim 1, characterized by: manually defined navigation zones. by determining the midpoints of opposite sides, those midpoints forming the main navigation lines between them and pre-determining the main navigation lines 11 Creating a graph that segments the data according to a defined step size. It includes the module.

4. A system compliant with Claim 3, characterized by: assigning a regional ID and line to each segment. a single graph in a hierarchical structure containing identity and segment order components It includes a graphics generation module that assigns the identity.

5. A system that complies with Claim 1, characterized by the fact that each robot has its own existing graphics. k user-defined forward steps along the route, starting from the identity The set of graphic IDs belonging to a graphic node and the words belonging to different robots if the sets in question have at least one common graph identifier It includes a collision and congestion analysis module that identifies potential collisions.

6. A system that conforms to Claim 1, characterized by its lateral orientation within the navigation area. its capacity is the sum of two robot widths and three minimum safety distances Equal to or greater than the navigation zone width a line structured to be evaluated based on the condition of being small The scrolling mechanism includes a rescheduling mechanism.

7. A system that conforms to Claim 1, characterized by its lateral offset of the main navigation line. the amount of lateral free space in the navigation region and in the system beforehand Line shifting that determines the shift within the defined maximum shift limits. It includes a rescheduling mechanism.

8. A system that complies with Claim 1, characterized by having the same system for each navigation zone. the number of robots allowed at a time, whether it's a single robot or multiple robots It includes regional capacity and policy definitions that enable its designation.

9. A system that complies with Claim 1, characterized by the presence of robots involved in the collision. calculates the Euclidean distance between the positions and the target points and the target. Prioritization that identifies the robot with the shorter distance as the winner. It involves a mechanism. 12 10. A system that complies with Claim 1, and its feature is that the winning robot is predefined. The robot includes a prioritization mechanism that determines priority according to a priority rule.

11. This is a system that complies with Claim 1, and its feature is that it determines the robot as the winner. the command to continue with the current plan and for the robot designated as the loser It includes a central decision-making unit that generates wait or parking commands.

12. A system that conforms to Claim 1, characterized by being controlled by a central decision-making unit. the transmitted continue, reschedule, wait or park command robot local control, which implements local path planning and motion control of the robot. and includes a planning unit. 13