Autonomous vehicle management system and method

Abstract

To provide an autonomous travel vehicle management system and a method thereof for dynamically managing a driving routine of one or a plurality of autonomous travel vehicles V included in a cooperation type system.SOLUTION: An autonomous travel vehicle management system comprises: one or a plurality of autonomous travel vehicles V communicatively connected to an external control center 300; and a route planner 302. In addition, in order to enable a dynamic expected safe area allocation mechanism, the autonomous travel vehicle management system is further configured to: continuously determine a change in an environment of the system; detect potential and existing hazards; and calculate and allocate adaptive emergency routes and associated exclusive safety areas to the autonomous travel vehicle V according to the hazard detection, and as a result, the autonomous travel vehicle management system will create an efficient and reliable system environment.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present application relates to an autonomous vehicle management system and method for dynamically managing movement routines of one or more autonomous vehicles.

Background Art

[0002] In recent years, due to the ever-increasing need to provide rapidly controllable work routines while addressing the decline in the working population, demand for human-cooperative autonomous robot systems has increased significantly. In such complex ecosystems, ensuring safety for both human workers coexisting in a predefined space and each of the autonomous components, while optimizing productivity, has become the top priority. As a result, modern conventional cooperative systems execute a large number of safety strategies implemented within the management core of the system.

[0003] Here, one common and effective strategy for further improving safety standards in cooperative systems has been found by dynamically allocating exclusive safety areas to respective human workers and machine workers included in a predefined work space, and implementing a safety stop interaction by the machine worker whenever two of these exclusive areas overlap.

[0004] As one of the most prominent examples, Japanese Patent Laid-Open Publication No. 2021-135801 applies a similar exclusive safety area mechanism to a system that manages the control of autonomous forklifts that pick up one or more pallets at a starting position and transport them to a predetermined location. The document specifies that when a forklift arrives at a given starting position, a predetermined system component allocates an exclusive safety area to the corresponding forklift based on the detected position of the forklift itself and the position and orientation of the required pallet. Furthermore, when providing an exclusive safety area to the aforementioned forklift, different allocation patterns are defined in order to further align the corresponding exclusive area allocation process with the requirements of the underlying management system. [Overview of the project]

[0005] Specifically, in a first embodiment of the allocation pattern described above, the system in the cited document specifies allocating a fixed, exclusive safety area to each existing forklift, large enough to cover all potential movements and / or emergency routes, including cases where a corresponding forklift is required to deviate from its originally assigned route, such as in the case of an emergency collision event. In addition, in a second embodiment, the allocation of such safety areas different from the initial movement route should only be performed after determining the need for additional emergency routes.

[0006] For this reason, while each of the above allocation patterns can further improve safety or efficiency within existing cooperative systems to some extent, none of the existing safety area allocation strategies still have the problem of simultaneously and sufficiently and synergistically improving both of the required system characteristics. For example, the first embodiment of the allocation strategy cited above can cover all potential emergency routes and thus provide an improved safety structure for all individuals present in a given system, but allocating all potential routes also requires allocating a large exclusive safety area to each machine unit, even if not in use, thus further restricting the movement of other system entities and ultimately leading to a decrease in system efficiency. On the other hand, the limitation of a small exclusive safety area as defined by the second embodiment allows for more diverse movement patterns for each existing machine, but allocating safety areas only in determined emergencies can result in a less safe environment, and at the same time, each machine is forced to spend a waiting period until a suitable emergency route and associated safety area are found by the system.

[0007] Therefore, in order to solve the above-mentioned problems and to further improve both efficiency and safety, which are clearly lacking in the latest cooperative autonomous systems, an autonomous vehicle management system and method according to the claims of the present invention are proposed herein. Furthermore, the dependent claims define preferred embodiments of the claimed invention.

[0008] For this reason, the autonomous vehicle management system according to the claimed invention may be configured to synergistically improve the safety and efficiency properties in a cooperative environment by specifically providing a dynamic safety area allocation strategy for one or more autonomous vehicles in the system, and in particular by implementing an expected hazard detection mechanism that works in conjunction with a dynamic and responsive vehicle path and exclusive area allocation strategy.

[0009] For this purpose, the autonomous vehicle management system of the claimed invention may be able to dynamically manage the driving routines of multiple autonomous vehicles contained in a predefined environment, such as a warehouse, a public or isolated street or road, or any given area that can be assigned as a place for a cooperative autonomous system. Thus, the invention is not limited to a particular autonomous system or a particular vehicle type, but can potentially be implemented in any autonomous system, including a closed automated supply system such as the automated forklift environment presented in the cited literature, as well as in outdoor structures such as tracked or wheeled self-propelled vehicles, buses, or even rail elements.

[0010] Overall structure of the autonomous vehicle management system With respect to the elements provided by the proposed invention, the proposed autonomous vehicle management system, according to a preferred embodiment, may first include at least an external control center which includes at least an exclusive area manager for managing exclusive safety areas that are present in or to be implemented within the system, and one or more autonomous vehicles are communicably connected to the external control center and at least a route planner which manages different travel routes assigned to one or more autonomous vehicles.

[0011] Here, the external control center may be configured as a centralized or distributed communications device permanently or restrictively coupled to each of the one or more autonomous vehicles by communication connectivity means such as Wi-Fi, Bluetooth®, or any other preferred wireless connectivity, which allows the external control center to transfer data and / or instructions to one or more autonomous vehicles and to receive information from one or more autonomous vehicles for internal processing. As a result, the exclusive area manager of the external control center may also be preferredly designed as a physical or computing module of the external control center and thus capable of communicating with one or more autonomous vehicles present in the system.

[0012] Furthermore, with respect to the route planner, each of its physical and / or computational elements may be equally implemented as modules in an external control center and thus share the same communication capabilities as those assigned to the exclusive area manager described above. In contrast, in an additional equally preferred embodiment, the route planner may also be designed as a standalone device having the same communication characteristics as those given to the external control center, or the route planner may be specifically implemented as an integrated component of each of one or more autonomous vehicles.

[0013] In addition, with respect to the initial characteristics of each of the aforementioned system elements, the route planner and exclusive area manager may be preferentially defined as connected centralized system components capable of defining and overseeing the current travel route and associated exclusive safety area assigned to each of one or more autonomous vehicles, based on one or more implemented mechanisms described later.

[0014] Calculation and allocation of initial routes and exclusive safety areas. For this reason, in a first preferred embodiment, the route planner is first preferentially configured to calculate and assign to each of one or more autonomous vehicles at least an initial travel path (hereinafter referred to as the "initial path"), which includes, for example, travel instructions used to define a preferred travel path and instruct a given autonomous vehicle to move within the system environment, while the exclusive area manager may further calculate and store at least an exclusive safety area, assigned to each of the generated initial paths, which defines areas where a given autonomous vehicle may be configured to automatically pause if it is found that an object is entering the respective safety area. As a result, using the above initial definitions of the initial paths and associated exclusive safety areas, the corresponding autonomous vehicle management system may generate an initial travel trajectory to which a given autonomous vehicle is permitted to move, in particular, so as to satisfy the current task requested by the vehicle (e.g., transporting a given element from a starting point to a destination) while simultaneously maintaining safety within the system environment using additionally provided exclusively defined safety areas.

[0015] Here, the generation and allocation of each initial route and exclusive safe area provided by the route planner and exclusive areas may vary depending on the internal structure of the comprehensive cooperative system. In the first embodiment, it may be preferable that a given task, such as the transport of a predefined object, be assigned to one of one or more autonomous vehicles included in the system (e.g., by an external message provided by the user or the system itself), and as a result, the autonomous vehicle is required to receive route instructions to safely travel from a predetermined starting position to its designated destination. In this embodiment, each autonomous vehicle may be able to send a request to the route planner to find a sufficient route to reach each, and the route planner may accordingly be configured to calculate the respective initial route for the corresponding autonomous vehicle based on the information contained in the received route calculation request (for example, the calculation request may include a given starting location, an ending location, a predetermined time by which the autonomous vehicle must reach the ending location, or a speed specification that the route planner may use to derive the initial route). In addition, after calculating a sufficient number of routes, the route planner may more preferably be configured to send the calculated initial routes to an exclusive area manager, which then calculates and assigns an exclusive safe area corresponding to the calculated initial routes.

[0016] The exclusive area manager may then be configured to calculate, or derive, a preferred, sufficient exclusive safety area for each initial route based on the information received along with the route planner's request and / or internal guidelines predetermined by the autonomous vehicle management system. As an example, the exclusive area manager may derive a corresponding exclusive safety area associated with a given initial route by allocating a predetermined area around the initial route having a predetermined size, the size of which may depend on a preset size parameter within the respective system. Furthermore, in addition to or instead of the above-described measures, the exclusive area manager may also be configured to assign additional parameters to further define the size or shape of the corresponding exclusive area.

[0017] Here, as an example, the exclusive area may be further configured to define an exclusive safety area in a time or place-dependent manner, preferably. For example, a given exclusive safety area may also be defined as a given area surrounding the current position of an autonomous vehicle assigned to an initial route, and the size of the area can again be predetermined by system-specific parameters. In this way, it is particularly possible to provide the most significantly efficient safety area determination process, as the remaining unused areas in the system environment can still be assigned to different system elements by allocating only the area currently surrounding each autonomous vehicle. Furthermore, in another example, the size or location of the exclusive area may be defined to vary depending on the current time, place, or environment or other parameters of the assigned autonomous vehicle, and as a result, the exclusive safety area can likewise be specifically adapted to different requirements included in the system as needed (e.g., different locations in the system require different safety distances).

[0018] As a result, based on the generation mechanisms described above, performed by the route planner and their respective exclusive area managers, it may be possible for one or more autonomous vehicles included in the system to determine a sufficient initial route and safety area required to reach a given predetermined location.

[0019] At the same time, it should be noted that the above-described exemplary determination processes for both the route planner and the exclusive area manager define only one of several possible embodiments in which the corresponding initial route and associated exclusive safe area may be generated, and therefore should not be considered to limit the general scope of the invention.

[0020] Therefore, as an alternative embodiment different from the above description, the generation and assignment of a given initial route should not be limited to, for example, a single autonomous vehicle present in the corresponding system, but can also be used to calculate a single initial route and assign it to multiple existing autonomous vehicles. This may be useful, for example, in the case of repetitive delivery routes or locally limited environments where the majority of autonomous vehicles are compelled to share, by definition, the same transport track, at least partially, because calculating a single initial route and assigning it to multiple autonomous vehicles can efficiently save the resources required to assign each of the autonomous vehicles its own seemingly identical initial route. The calculation and definition of each exclusive safety area can then also be performed in a vehicle-specific manner, for example, by defining each of the exclusive safety areas to be used depending on the location of the vehicle to be assigned, or conversely, by broadly determining a predefined area surrounding the initial route shared by all the corresponding autonomous vehicles.

[0021] In addition, the existence of initial routes and exclusive safety areas already calculated and assigned to other autonomous vehicles may also be used by the route planner and exclusive area manager to determine and calculate corresponding new initial routes.

[0022] Here, for example, after successfully calculating and allocating each initial route and the exclusive safety area associated with each initial route, the external control center may be further configured to store the information regarding the calculated and allocated initial routes and associated exclusive safety areas in a predetermined storage area, preferably provided by an internal or external storage element connected to the external control center, such as a hard drive, a storage server, or any other entity capable of reliably storing information generated or received by the external control center. Based on this, the route planner and / or exclusive area manager may, for example, also preferably, refer to the stored information regarding initial routes and associated exclusive safety areas that are already in use by multiple other autonomous vehicles present in the system, with respect to the calculation of initial routes or associated exclusive safety areas, and calculate new initial routes or associated exclusive safety areas based on the referenced information.

[0023] As a result, in one preferred embodiment, the route planner may also be configured to extract information from a predetermined memory area regarding the current or expected location of existing initial routes and / or exclusive safety areas already assigned to different autonomous vehicles in the system, and to calculate a corresponding initial route in a manner that avoids contact with or proximity to the other autonomous vehicles. For this purpose, for example, the route planner may preferably be configured to calculate only initial routes that do not overlap with stored initial routes or exclusive safety areas already assigned to any of the other autonomous vehicles, thereby efficiently avoiding potential collisions with other elements in the system. Similarly, the exclusive area manager may also be configured to calculate and assign an exclusive safety area to a calculated initial route only if the safety area does not match an initial route or exclusive safety area already calculated and assigned to a different vehicle, so that the generation and assignment of the corresponding safety area can also be performed in a manner that promotes safety.

[0024] Therefore, specifically, by providing a customizable and scalable route generation mechanism, a more accurate and therefore safer route management process can be made available, thus generating improvements over current state-of-the-art management systems based on the aforementioned mechanism that adaptively calculates initial routes and associated exclusive safety areas and assigns them to predefined autonomous vehicles.

[0025] Behavior of one or more autonomous vehicles Furthermore, following the calculation of the initial route and associated exclusive safety area assigned to at least one of the one or more autonomous vehicles present in the system, the exclusive area manager of the external control center may preferably transfer the initial route and associated exclusive safety area (including movement instructions) to the assigned autonomous vehicle.

[0026] To this end, the external control center may keep track of the current initial routes and exclusive safety areas currently allocated to existing autonomous vehicles in the system by initially and additionally storing information relating to the initial routes and exclusive safety areas in a predetermined storage area, such as a storage area used to calculate the initial routes and exclusive safety areas, or another storage area different therefrom. At the same time, the stored information may also preferably further include further specifications or instructions connected to the stored initial routes and the associated exclusive safety areas, such as specifications of the allocated autonomous vehicles (e.g., registration numbers, current locations, initial requests transferred to the vehicles, etc.), or dynamically updated parameters such as the current position of the allocated autonomous vehicles, so as to enable an accurate review of the vehicles' actions when necessary.

[0027] Each one of the one or more autonomous vehicles allocated to the calculated initial route and the associated exclusive safety area may then preferably receive the respectively calculated initial route and exclusive safety area from the exclusive area manager, and is further configured to travel from a predetermined starting point to a destination point based on the allocated and received initial route.

[0028] Here, as mentioned above, the information received for the initial route and the exclusive safety area may include additional specifications for the corresponding autonomous vehicle that indicate how the autonomous vehicle is required to travel. By way of example, the initial route may include additional information such as a start time, a start location, and / or a speed required for the allocated autonomous vehicle to perform the travel. Furthermore, additional environmental information or conditions may equally be included in the initial route or exclusive safety area information, such as specific stop specifications (e.g., in the case of a detected object such as a detected "stop" sign or a red traffic light) that require the vehicle to stop, such that each of the one or more autonomous vehicles present in the system may receive specific instructions on how to perform its required action.

[0029] Furthermore, with regard to the general composition of existing autonomous vehicles included in the system, the design of each autonomous vehicle is not limited to a specific task or superstructure, and can be generally assigned to any autonomous vehicle layout capable of self-propulsion based on received instructions, in particular, such as the aforementioned initial route and exclusive safety area. Therefore, in the context of the present invention, an "autonomous vehicle" may preferably be initially understood as any movable entity that is at least compatible with the mechanisms of an additional external control center included in the system and can automatically comply with the aforementioned instructions.

[0030] Hazard detection mechanism Therefore, by using the initial mechanism of an external control center and one or more autonomous vehicles to calculate the initial route and the associated exclusive safety area, and assign the same to each of the one or more autonomous vehicles in the system, the present invention can sufficiently and dynamically manage the allocation of intended movement routines even for a plurality of autonomous vehicles moving simultaneously within a predefined system space.

[0031] At the same time, the proposed autonomous vehicle management system further comprises an additional emergency route mechanism associated with an emergency safety area calculation to equally cover the generation and specification of potential emergency routes for each autonomous vehicle required in a cooperative system, so as to avoid collisions with unexpected objects such as human workers that coexist in the system environment and ignore the currently allocated exclusive safety area, which may further improve the overall system safety and efficiency.

[0032] Specifically, for this purpose, the autonomous vehicle management system may preferably adopt an additional expected hazard detection mechanism coupled to an alternative route (hereinafter also referred to as a "candidate route"), and an exclusive safety area calculation process that can enable an additional dynamic adaptive collision avoidance strategy.

[0033] For this reason, in order to initially generate such additional anticipated hazard detection mechanisms, one or more autonomous vehicles present in the external control center and / or system may be further configured to detect changes in the environment around the initial path assigned to at least one of the one or more autonomous vehicles.

[0034] Here, “detection of environmental changes” may, preferentially, mean any potential recognition, including optical, electrical, acoustic, or any other type of recognition available for identification, used in the field where an external control center or at least one of one or more autonomous vehicles recognizes objects or other entities (e.g., humans or mobile machines) within the system and detects changes in object position or other detectable characteristics for hazard determination.

[0035] Accordingly, in a first preferred embodiment, for example, one or more autonomous vehicles of the system may be further equipped with one or more detectors, such as onboard cameras, infrared sensors, or position scanners, which can detect objects or entities present in the system's environment and can at least track the location or other properties of detected objects over a certain predetermined amount of time. Similarly, an external control center may also be equipped with such detectors, exemplary by positioning the detectors at predetermined locations in the system's environment and connecting them communicably to the external control center, so that each of the elements present in the current autonomous vehicle management system may be used as a detection component utilized in the anticipated hazard detection mechanism. Furthermore, in additional embodiments, the detection of a particular object may also be based on the reception and analysis of additional information detected outside the system itself, such that each detection mechanism performed here is not limited to active "detection" performed only by one or more entities of the corresponding system, but can be similarly understood as a collaborative detection system, by exemplary using GPS data of a particular object recognized by an external satellite system, for example, and transferring the data to the external control center or at least one of the one or more autonomous vehicles.

[0036] As a result, in order to establish the above-described anticipated hazard detection mechanism in the present invention, an external control center and / or one or more autonomous vehicles configured to detect corresponding environmental changes around an initial path may be configured to recognize objects present around an assigned initial path by observing a predetermined area around a given initial path using at least one of the above-described detection mechanisms, and to detect time-dependent behavior, i.e., preferably changes in the spatial position, velocity, or any other desirable properties of each of the so-called recognized objects around the environment.

[0037] Furthermore, each external control center and / or one or more autonomous vehicles performing object detection may be further configured to utilize the environmental information thus generated for the anticipated hazard detection mechanism and, therefore, to enable the adaptive and dynamic generation of emergency routes for each of one or more autonomous vehicles, to determine, based on the detected behavior of the recognized objects, whether one or more of the recognized objects should be identified as a potential threat to one or more autonomous vehicles assigned to the observed initial route, and in response to the determination, the external control center and / or one or more autonomous vehicles may also be further configured to calculate appropriate emergency routes and associated exclusive safety areas to be used by one or more autonomous vehicles if, at some point, the respective object becomes a real, i.e., actual hazard. Therefore, compared to existing emergency area allocation methods currently used in state-of-the-art management systems, the emergency route and safety area generation mechanism of the present invention may be particularly configured to perform a proactive, adapted calculation of emergency routes and associated exclusive safety areas whenever a given object has already been identified as a potential hazard (i.e., an object that may potentially pose a hazard to one or more autonomous vehicles in the future) rather than an actual (i.e., current) hazard.

[0038] As a result, specifically, each of one or more autonomous vehicles is equipped with multiple corresponding emergency (route) strategies already in use (when faced with a threat), and at the same time, the amount of required emergency routes and associated exclusive safe areas can be significantly reduced using the adaptive nature of the correspondingly calculated emergency routes. Therefore, based on the adaptive emergency route and associated safe area generation mechanism described above, the potential waiting time required by the corresponding autonomous vehicle to calculate a functional emergency route when faced with a real hazard can be efficiently avoided.

[0039] Based on this, the corresponding hazard detection and adapted emergency route calculation and associated safety area calculation strategies performed by the proposed autonomous vehicle management system may be processed with the following priorities:

[0040] In the first step, an external control center and / or one or more autonomous vehicles, which detect changes in the environment around a predefined initial path, may first select information about recognized objects and then determine the hazard of each object based on this information.

[0041] To this end, an external control center or one or more autonomous vehicles may be configured to collect object-specific information from its own storage area or to request necessary information from other entities of the autonomous vehicle management system, for example, by preferentially and continuously storing information such as the location, time, or even speed of each object recognized in an object-specific manner, and then retrieving this information for hazard detection. Here, as an example, in contrast to individual storage areas, each external control center or one or more autonomous vehicles may be connected to a central storage entity, such as a storage server or a communicably connectable hard drive, where each of the detected object information may be stored and accessed sequentially and alternatively.

[0042] Next, if each external control center or one or more autonomous vehicles successfully reads information about a given detected object, the determination of the hazard level of the given object (also called the “hazard determination process”) may preferably be performed by classifying the object into one of several hazard levels defined by the system based on the above information.

[0043] In a preferred embodiment, the autonomous vehicle management system may, for example, define the hazard of each detected object and include at least three hazard levels, which may be further subdivided by the names "actual hazard," "potential hazard," and "no hazard," as described later. Specifically, an object classified as an "actual hazard" may be defined as an object that appears to collide with or otherwise harm each autonomous vehicle within a foreseeable timeframe, requiring an immediate response from at least one of the autonomous vehicles. An example of such an object might be a person jumping in front of a moving vehicle. A “potential hazard,” in contrast, may be seen as an object that could potentially become an “actual hazard” at some point, but is not yet perceived as an actual threat to one or more of the system’s autonomous vehicles. An example of this might be a human worker walking next to a moving autonomous vehicle, but still quite far away from it. Finally, an object classified as "not dangerous" may be any object that, at the time of detection, is not considered to pose any potential threat to the autonomous vehicle, at least in the near future. This may occur, for example, if a given object is maintaining a sufficient distance from each of the vehicles, and therefore, in effect, there is no possibility of harm from that object in the immediate future.

[0044] In this process, the classification of each recognized object into one of the aforementioned danger levels may preferably be performed by comparing object information received by an external control center or one or more autonomous vehicles with one or more parameters and / or thresholds predetermined by the system, which are defined by one or more existing autonomous vehicles.

[0045] For this reason, in a first preferred embodiment, the determination of the danger level of a recognized object may be performed, for example, by analyzing the distance of the recognized object to each of one or more autonomous vehicles currently moving along an initial path that is present in or observed within the system, and comparing the distance with a predefined distance threshold that serves as a categorization upper or lower limit.

[0046] Therefore, in a given embodiment, an external control center or one or more autonomous vehicles may be configured to, for example, preferably extract at least the location, most preferably the current location, of each object from the corresponding object information, and calculate the distance of the object to each of the corresponding vehicles, in order to determine the danger of each object. Subsequently, the external control center or one or more autonomous vehicles may further compare the calculated distance with a predetermined distance threshold and assign one of the above danger levels to the object for each autonomous vehicle associated with a given distance.

[0047] As a result, assuming that the distance of each object to one of one or more autonomous vehicles is determined to be, for example, below a predetermined potential hazard value but greater than a predetermined actual hazard value, the external control center and / or one or more autonomous vehicles may be configured to define the object as a “potential hazard” (for the autonomous vehicle assigned to the distance used), and distances smaller than a predetermined actual hazard value may be associated with categorization as an “actual hazard” for a particular vehicle. In contrast, distances greater than the aforementioned potential hazard value may be associated with categorization as “no hazard.”

[0048] As a result, based on the aforementioned vehicle-specific determination of hazard levels performed for each object and each autonomous vehicle present in the system, the present invention can provide a mechanism that can reliably predict the hazard of any object recognized by an external control center and / or one or more autonomous vehicles, and can implement effective measures to avoid vehicle collisions and / or waiting times even before detecting actual hazards present in the system. In addition, since each identified hazard level is similarly already defined based on a specific vehicle distance and therefore always assigned to each vehicle, each calculated hazard level can also be viewed as an efficient method that further improves the adaptability and efficiency of the system, in particular by performing emergency route generation only for specific vehicles in the case of a specific hazard level, thereby significantly reducing the number of safety area allocations required and the calculation procedures generally required in the system.

[0049] Furthermore, apart from the distance conditions described above, other object information and / or system thresholds may also be used preferentially to create and / or influence alternative or additional risk level categorizations.

[0050] For example, in another embodiment, additional characteristic properties of a given object, such as the type of object (e.g., whether the object is a person or a machine), its size, or its past trajectory, may also be included in determining the respective risk levels.

[0051] In addition, in a further preferred embodiment, an external control center or one or more autonomous vehicles may also be equally configured to determine a given risk level not only by parameters that indicate the current nature of the object and the corresponding autonomous vehicle, but also by further relying on extrapolations of already existing movement patterns of both the object and the corresponding autonomous vehicle that may potentially exist in the near future.

[0052] For this reason, in an additional preferred embodiment of the risk assessment process described above, the external control center and / or one or more autonomous vehicles may be equally configured to determine, based on information (such as trajectories or time-dependent sequences of location information) that references an object and the corresponding autonomous vehicle, predict the distance between the object and the corresponding autonomous vehicle at a given estimated time, and then categorize the risk level of each object based on the predicted distance. In this way, even if more reliable emergency detection can be generated, potential threats (and effective interventions, respectively) to one or more existing autonomous vehicles in the system may be detected even earlier.

[0053] Vehicle behavior based on hazard detection In addition, following the identification of the respective hazard levels assigned to each recognized object and associated with each corresponding autonomous vehicle, i.e., the determination of potential and actual hazards of concern at a given point in time of recognition, the autonomous vehicle management system may further implement several different measures based on the determined hazards, including generating emergency routes and assigning associated safe areas for each of one or more autonomous vehicles associated with a given hazard level, in order to effectively overcome any threats currently present within the system.

[0054] Based on this, at least one or more of the following criteria may be implemented, preferably by the autonomous vehicle management system or its corresponding elements, in which case, for the sake of better understanding, each of the criteria may be implemented, as herein referred to, exemplary for the first autonomous vehicle (hereinafter referred to as the "first autonomous vehicle"), which represents one or more autonomous vehicles of the system that are assigned to an initial route (and most preferably already moving along the initial route) and associated with one or more hazard level detections determined in a previous hazard assessment process. Therefore, if it is found that the first autonomous vehicle is associated only with a “no hazard” level determination, that is, if no potential and / or actual hazards can currently be found by the hazard determination process for the first autonomous vehicle, the autonomous vehicle management system may be configured to suspend any additional actions relating to the first autonomous vehicle and thus allow the first autonomous vehicle to continue moving along the initial path that was initially assigned. Furthermore, if the first autonomous vehicle may be assigned to at least one “potential hazard” determination, and therefore at least one potential hazard has been detected for the first autonomous vehicle, the autonomous vehicle management system, more specifically the external control center and / or one or more autonomous vehicles, may be configured to perform a route candidate generation process that defines the generation of potential emergency routes to be used by the first autonomous vehicle when an object pointing to a “potential hazard” becomes an actual hazard at some point. To this end, the external control center and / or one or more autonomous vehicles may be configured to preferentially compute and store one or more potential emergency routes associated with the first autonomous vehicle (hereinafter also referred to as “route candidates”) that can avoid the hazard generated by each recognized object, and further compute and store an exclusive safe area (hereinafter referred to as “safety area candidate”) associated with each of the route candidates. As a result, if a previously recognized potential hazard can change from a potential hazard to an actual hazard, the potential emergency routes may already be available to the first autonomous vehicle. In addition, if a given object is determined to be an "actual hazard" to the first autonomous vehicle, and therefore indicates at least one "actual hazard" determination event in the corresponding hazard determination process, the first autonomous vehicle may be configured to perform a route switching process to efficiently avoid each of the hazards currently imminent in the vicinity of the first autonomous vehicle. To this end, the first autonomous vehicle may preferably select one of the previously calculated and assigned already stored route candidates, set the selected route candidate as a new initial route to follow, and the exclusive safety area candidate associated with the selected route candidate as a new exclusive safety area, and finally continue moving along the new initial route to continue effectively moving to the destination without requiring any further steps. In contrast, if the first autonomous vehicle is unable to select a route candidate and / or associated safety area candidate, the first autonomous vehicle may preferably suspend its movement to result in forced collision avoidance by the autonomous vehicle management system.

[0055] As a result, an efficient predictive hazard avoidance strategy may be generated based on the process steps described above, which are performed by the first autonomous vehicle in the event of a detected hazard level. In particular, since the route candidate generation process described above is usually performed before the route switching process (specifically, the system can always detect potential hazards earlier than actual hazards based on the hazard detection mechanism described above), the emergency route generation and usage strategy described above may be able to provide the first autonomous vehicle with the possibility of an emergency route that is always required, even before the actual detection of an actual hazard, which leads to the effect of effectively avoiding potential waiting time for calculating an emergency route when necessary (i.e., at the time of actual hazard detection).

[0056] Simultaneously, the above-described process steps enable improved efficiency and continuous allocation of required exclusive safety areas within the system. Thus, since each of the above steps relies on previous detections of corresponding hazard levels assigned to a given (first) autonomous vehicle and at least one object, preferably performed in a continuous manner within the proposed system as described above, the above-described process steps may also preferably be performed continuously, for example, in a predetermined sequence of consecutive times, to effectively adapt to the occurrence of changes detected within the system environment. Similarly, one or more of each step may also preferably be processed in parallel if two or more of the above conditions are met at the same time, for example, if two or more objects are determined to cause a “potential hazard” or “actual hazard.” In this way, an external control center and / or one or more autonomous vehicles may be configured to calculate and assign a given one or more route candidates and associated safety area candidates to the first autonomous vehicle associated with the first object, for example, while the first autonomous vehicle itself may simultaneously perform a route switching process for a second object associated with the “actual hazard” detection. Furthermore, several route candidate generation processes and / or route switching processes may be executed simultaneously.

[0057] Route candidate generation process The following may provide further information and / or description of possible embodiments of a route candidate generation process performed by an external control center and / or one or more autonomous vehicles of the proposed system, after performing hazard detection and generating different hazard levels associated with the recognized object and a given autonomous vehicle. Here, if a given process step can be assigned to any single autonomous vehicle among one or more autonomous vehicles present in the system, it will also be referred to as the first autonomous vehicle.

[0058] Therefore, as described above, each route candidate generation process may preferably be performed by an external control center and / or one or more autonomous vehicles when there is a change in the system environment, i.e., preferably when a given object is recognized as a “potential hazard” to a given first autonomous vehicle in light of a hazard detection process preferentially performed by the system. In this regard, each route candidate generation process may primarily aim to generate one or more emergency routes for the first autonomous vehicle, also referred to below as “route candidates,” to efficiently avoid threats such as collisions between the first autonomous vehicle and objects when the change in the environment may at some point become an “actual hazard,” and at the same time, relevant safety area candidates may be assigned to the route candidates to enable an adaptive hazard-dependent safety area allocation mechanism.

[0059] For this reason, in a first preferred embodiment of each route candidate generation process, the generation / calculation of one or more route candidates may be performed preferentially before the calculation and allocation of the associated safe area candidates.

[0060] Here, one or more route candidates may be calculated based on a predetermined route candidate calculation algorithm, which is implemented in an external control center and / or one or more autonomous vehicles and supplied with the aforementioned information to determine different arbitrary route candidates. Specifically, this includes object-specific information relating to several environments, i.e., the properties of objects determined to be "actual hazards," system internal information including vehicle-specific characteristics, and information relating to several environments, i.e., object-specific information relating to the properties of objects determined to be "actual hazards."

[0061] Based on this, in a preferred embodiment, the calculation of one or more route candidates may include, for example, the calculation of at least first and second route candidates for a first autonomous vehicle, preferably the first and second route candidates being different from the initial route of the first autonomous vehicle, and the first and second route candidates may be calculated by implementing object information and system internal information into the first and second route candidate calculation algorithms.

[0062] For this reason, the first route candidate calculation algorithm may be implemented such that, for example, it is calculated for a first autonomous vehicle, and each first route candidate output by the first route candidate calculation algorithm continuously maintains the maximum distance of the first autonomous vehicle to a potential hazard represented by an associated object. Accordingly, as an example, the first route candidate calculation algorithm may be used to generate first route candidates by determining, preferably, alternative routes for the first autonomous vehicle to travel, which are either away from the location of each potential hazard or away from the location where each potential hazard could actually become a hazard.

[0063] In contrast, the second route candidate may be defined in the opposite way as the output of the second route candidate calculation algorithm, such that the first autonomous vehicle maintains a predetermined distance from potential hazards, and the second route candidate may, if possible, be realigned with the initial path of the first autonomous vehicle. Here, as an example, in order to realize the second route candidate, the second route candidate calculation algorithm may preferably be configured to further form the second route candidate by implementing an evasion curve in the second route candidate to avoid potential hazardous collision events, and such that the second route candidate subsequently reconnects to the initial path assigned to each initial path.

[0064] Therefore, by providing a second route candidate, a potential emergency route may be generated that allows for minimal change to the vehicle's initial route and thus leads to an efficiency-oriented route alternative, while the first route candidate may be defined to ensure maximum safety for all entities in the system.

[0065] Furthermore, with respect to possible information that an external control center and / or one or more autonomous vehicles may need to calculate one or more route candidates, preferably using at least one of the route candidate calculation algorithms described above, the external control center and / or one or more autonomous vehicles may be configured to receive specific information about objects associated with potential hazards and a first autonomous vehicle for which one or more route candidates are calculated, and to calculate each route candidate based on different predetermined relationships that can be extracted from the above information.

[0066] As an example, in a preferred embodiment, an external control center and / or one or more autonomous vehicles may calculate at least one route candidate based on the relative position of a corresponding object with respect to a first autonomous vehicle, the relative position may define one of the relationships described above, and can be extracted by the external control center and / or one or more autonomous vehicles by receiving position information of the corresponding object and the first autonomous vehicle for route candidate calculation. To this end, the external control center and / or one or more autonomous vehicles may be configured to implement an internal relationship conversion module that can read such information by accessing a memory area that currently stores information about the detected object and the currently used autonomous vehicle, such as the memory area mentioned specifically for initial route calculation and hazard detection, and then calculate the relationships (based on the read information) necessary for each route calculation using the read information.

[0067] Based on this, a preferred route candidate calculation process, based on the relative positions of the object and the first autonomous vehicle as exemplary relational parameters, may result in different route orientation candidates depending, for example, on which side of the first autonomous vehicle the object is currently located. For example, if each object is identified as being on the left side of the first autonomous vehicle, the external control center and / or one or more autonomous vehicles may be configured, based on one or more underlying route candidate calculation algorithms, to calculate / determine a suitable route candidate oriented to the right side of the first autonomous vehicle to ensure avoidance of any potential threats caused by the corresponding object. Conversely, the same technique may be applied similarly to other relative positions of the object, for example, if the object is on the right side, a route candidate on the left side may be preferred, or if the object is in front of the first autonomous vehicle, route candidates oriented strictly to the side or even behind the vehicle may be preferred.

[0068] As a result, given that the corresponding route candidate calculation process may take into account specific and predetermined relationships between each object and the first autonomous vehicle, each of the thus generated route candidates can be generated in an effective, specifically adaptive, manner, thus leading to more efficient generation of emergency routes, while simultaneously reducing the allocation of required route candidates and associated exclusive safety areas to provide each autonomous vehicle's respective safe movement routine.

[0069] In addition, in a further preferred embodiment, the calculation of the corresponding path candidates described above may be further supplemented not only by using information and / or relationships of the corresponding objects and the first autonomous vehicle regarding the current characteristics of the environment, but also by such information and / or characteristics that may be expected in the (near) future.

[0070] Therefore, in a given embodiment, the external control center and / or one or more autonomous vehicles may be equally configured to calculate and determine one or more route candidates based on predictions, i.e., predictions of object movement, in order to further improve the accuracy of each route candidate calculation output.

[0071] For this reason, in a preferred embodiment, an external control center and / or one or more autonomous vehicles, in particular the aforementioned route candidate calculation algorithm implemented therein, may also be configured to perform an initial prediction calculation process with respect to the corresponding route candidate calculation, for example by performing an extrapolation mechanism on object data, and then utilize the prediction output for the actual calculation of the corresponding route candidate, such as the predicted location of a given corresponding object at a given still prospective time. Thus, safety within the corresponding cooperative system may be further improved, in particular, since each autonomous vehicle may have multiple different emergency routes to consider for any potential behavior of the object in question. In addition, the accuracy and efficiency of each candidate calculation procedure may be further improved, specifically, since the continuous extrapolation analysis and output may enable an earlier and more accurate description of the object's future movement.

[0072] Calculation and allocation of relevant safety area candidates Furthermore, after successfully calculating one or more route candidates, the external control center and / or one or more autonomous vehicles may be configured to further calculate and store, for each of the calculated route candidates, a corresponding exclusive safe area, also referred to below as a "safety area candidate," primarily to ensure the continued safe movement of the first autonomous vehicle even while it is moving along one of the calculated route candidates.

[0073] Here, each candidate safety area may preferably have the same characteristics as an exclusive safety area preferentially associated with each of the initial paths assigned to one or more autonomous vehicles, and as a result, a candidate safety area may be seen as equivalent to an assigned safety area for the corresponding autonomous vehicle (here, the first autonomous vehicle), and under normal conditions, the movement of the corresponding autonomous vehicles should be ensured primarily by temporarily halting the movement of one or more autonomous vehicles (including the first autonomous vehicle) when it is found that an object has entered an assigned safety area (see further conditions below).

[0074] As a result, in the present invention, safety area candidates associated with a given route candidate may be calculated based thereon, and the parameters and conditions may, in a preferred embodiment, preferentially, be the same as those already described with respect to the exclusive safety area associated with a previous initial route. That is, if a given route candidate can be sufficiently calculated, an external control center and / or one or more autonomous vehicles may preferably calculate and assign a given safety area candidate associated with the route candidate by allocating a given area around the route candidate, including a predefined size, in accordance with the process of an exclusive area manager for exclusive safety areas, to further define the size or shape of the corresponding safety area candidate, including additional system or time-dependent parameters for safety area determination, or even further, to generate safety area candidates that move along the first autonomous vehicle while traversing the safety route candidate and have the same predefined size and shape.

[0075] Furthermore, after having fully computed each of the one or more route candidates and associated safety area candidates, the external control center and / or one or more autonomous vehicles may then preferentially store at least one route candidate and associated safety area candidate in a predetermined memory area (e.g., one of the potential memory areas also described with respect to the initial route calculation process), and assign each route and safety area to the corresponding first autonomous vehicle by, for example, assigning a vehicle-specific marker, such as a vehicle-specific ID, to the instructions thus generated. Thus, specifically, since each of the computed routes and safety areas may already be essentially organized within an existing system, the first autonomous vehicle may be able to efficiently read the information and instructions contained within the route candidates and associated safety area candidates.

[0076] In addition, in a further preferred embodiment, an external control center and / or one or more autonomous vehicles may be configured to preferentially store their respective route candidates and associated exclusive safety area candidates, and similarly add each of the stored exclusive safety area candidates to the area currently assigned to the exclusive safety area of ​​each first autonomous vehicle. Thus, it may be possible to further increase the efficiency and security of one or more autonomous vehicles in the system by adding each of the safety areas associated with potential emergency routes used by the autonomous vehicles to the exclusive safety area of ​​the vehicles, even before the actual act of using an emergency route, so that the area to which the autonomous vehicle is to move in the event of an emergency is already preferentially assigned, i.e., fixed, to the corresponding autonomous vehicle. At the same time, this eliminates the need to introduce additional security instructions to the autonomous vehicle, since the autonomous vehicle is already configured to pause its movement if a given object is located in the existing route candidate space, specifically, if the object is already located within the assigned exclusive safety area.

[0077] In addition, in a further preferred embodiment, the external control center and / or one or more autonomous vehicles may also be configured to further manage and / or update the stored route candidates and associated safety area candidates even after successfully storing each route candidate and associated safety area candidate in the corresponding memory area, thereby dynamically changing both the form of the currently used exclusive safety area and the availability of route candidates and exclusive safety area candidates used by the autonomous vehicles. To this end, the external control center and / or one or more autonomous vehicles may, for example, preferably, preferably, when certain conditions are met within the system, erase and / or update a given stored route candidate and associated exclusive safety area candidate, and potentially further adapt the exclusive safety area assigned to the autonomous vehicle associated with the erased and / or updated route and safety area according to the erase and / or update process performed.

[0078] For this reason, in one preferred embodiment, the external control center and / or one or more autonomous vehicles may be configured to clear and / or update each route candidate and associated safety area candidate whenever a given predetermined change in the environment is detected (e.g., a recognized object associated with a given hazard level moves, resulting in a change in the previous hazard level) and / or whenever a predetermined time (after storage) has elapsed. In addition, in another preferred embodiment, the external control center and / or one or more autonomous vehicles may also be configured to adapt the exclusive safety area currently used by the autonomous vehicle assigned to the cleared and / or updated route candidate or exclusive safety area candidate by similarly clearing and / or updating areas previously added to the exclusive safety area of ​​the autonomous vehicle, preferably similarly, based on the respective clearing and / or update. Here, as an example, the adaptation of a preferably existing exclusive safety area may, exemplary, include the erasure of an area associated with an erased exclusive safety area candidate that was previously added to the exclusive safety area of ​​the autonomous vehicle (for example, each of the stored exclusive safety area candidates may have been previously added to the vehicle's exclusive safety area after route candidate calculation). In another example, in the case of an updated, i.e., modified exclusive safety area candidate, the areas previously added to the exclusive safety area of ​​each autonomous vehicle may be changed to the updated characteristics of the exclusive safety area candidate. Thus, it may be possible to efficiently and dynamically update and align the areas assigned to each exclusive safety area present in the system, resulting in the effect of maintaining a fast and reliable response to a given change in the system's environment.

[0079] Internal pathway candidate selection mechanism Furthermore, in another preferred embodiment of the present invention, it may also be equally possible, preferably even before the storage step described above, for each route candidate and associated safety area candidate to similarly experience an additional route candidate selection mechanism, thereby further enabling the exclusion of invalid or even prohibited route candidates calculated by an external control center and / or one or more autonomous vehicles, and thus further improving the efficiency and accuracy of the underlying candidate calculation procedure.

[0080] For this reason, in an additional preferred embodiment, an external control center and / or one or more autonomous vehicles, where each route candidate and associated safety area candidate for the first autonomous vehicle has been calculated, may further be configured to perform an additional selection process on the calculated route candidates and safety area candidates, and then store only such route candidates and associated safety area candidates that have been permitted within the selection process.

[0081] Here, each selection process itself may be carried out by an external control center and / or one or more autonomous vehicles, preferably based on a plurality of selection parameters that further define the usefulness of a given route candidate and / or safety area candidate. For example, within each selection process, the external control center and / or one or more autonomous vehicles may preferably be configured to analyze predetermined characteristics of each route candidate and / or associated safety area candidate, and may be allowed to later store only such routes and safety areas where the characteristics satisfy predefined conditions implemented in the selection process. As an example, each selection process may include an analysis of the length of the route candidate, the time required for each first autonomous vehicle to travel along the route candidate, or the amount of safety area required for this route, and may allow subsequent storage of each route candidate and associated safety area candidate only if at least one, or preferably all, of the analyzed characteristics meet a predefined threshold.

[0082] As a result, the calculation of candidate routes and safe areas that may potentially fail to meet a predetermined set of standards in the underlying management system can be effectively excluded using the additional selection process described above, thereby further improving the efficiency and accuracy of the underlying system by effectively conditioning the allocation of emergency routes and safe areas.

[0083] In addition, the selection mechanism described above may preferably also be used to further align the calculated route candidates and associated safety area candidates with the initial routes and route candidates and safety areas already present in the system.

[0084] Specifically, in cooperative systems, security issues can arise when a given autonomous vehicle may share a portion of one of the initial routes, route candidates, or associated safety areas, primarily in cases where collisions between each vehicle may not be entirely avoidable. Therefore, each selection mechanism may similarly, preferably, be used to exclude calculated route candidates and associated safety area candidates that may fall, at least partially, into the system's already existing (i.e., used and / or stored) initial routes, route candidates, and / or associated safety areas.

[0085] For this reason, additional embodiments of the path candidate selection process described above may equally include a mechanism in which potentially uncertain path candidates and associated safe area candidates may be excluded before storage.

[0086] To this end, in a preferred embodiment, the external control center and / or one or more autonomous vehicles may further send an additional request to the external control center's exclusive area manager, primarily to cause the corresponding exclusive area manager to implement the safety inspection mechanism described above, and preferably, in response to the exclusive area manager's acceptance after the safety inspection, the external control center and / or one or more autonomous vehicles may be configured to store only the respective route candidates and associated safety area candidates when they receive a corresponding storage permission message from the exclusive area manager.

[0087] Here, the exclusive area manager's safety inspection mechanism may preferably be handled by the exclusive area manager in a manner similar to how the exclusive area manager may define the exclusive safety area of ​​the associated initial route described above. Specifically, the exclusive safety area may also preferably store and / or read information about an already existing, i.e., stored initial route, route candidate, and associated safety area candidate and / or exclusive safety area (preferably in a manner similar to that described above with respect to the exclusive safety area), and then preferably, based on the stored route and safety area information described above, send only the corresponding storage permission message to the external control center and / or one or more autonomous vehicles if the characteristics of each route candidate and associated safety area candidate sent to the exclusive area manager for safety inspection can satisfy a plurality of predetermined conditions.

[0088] For this reason, in a preferred embodiment of the safety inspection mechanism described above, the external area manager may be configured to read information, for example, currently stored exclusive safety areas and safety area candidate information present in the system, which may preferably be the current and / or time-dependent location of each safety area, compare this information with the characteristics of the safety area candidate to be inspected by the exclusive area manager, i.e., the calculated location, and only send a corresponding storage permission message to the external control center and / or one or more autonomous vehicles if each inspected safety area candidate does not overlap with any of the currently stored exclusive safety areas and safety area candidates.

[0089] As described above, even when a given autonomous vehicle needs to travel along one of the provided safe route candidates, it may be efficiently avoided that the given autonomous vehicle could collide with any other autonomous vehicle in the system. Specifically, any (even potential) dangerous convergence between two autonomous vehicles can be completely prevented, for example, by pre-selecting (i.e., pre-filtering) route candidates that share even a portion of the same safe area with another already existing initial route or route candidate.

[0090] In addition, the above-described mechanism may also enable further improvements in the adaptability of the entire autonomous vehicle management system.

[0091] For example, the route inspection mechanism described above may equally be used to further adapt and improve the entire process of assigning and using travel routes included in the provided system, rather than merely involving the determination and selection of new route candidates and associated safety area candidates stored for a given autonomous vehicle.

[0092] For this reason, in an additional preferred embodiment, the exclusive area manager implementing the security inspection mechanism described above may be configured not only to output (or not output) a given storage permission message to an external control center and / or one or more autonomous vehicles after detecting the sufficiency / insufficiency of the corresponding route candidate or associated safe area candidate, but the detection may also be used to further adapt existing (i.e., stored) route and safe area definitions or assignments to improve the overall efficiency of the management system.

[0093] Therefore, in additional embodiments, if the exclusive area manager detects during security inspection that, in some examples, a requested safety area candidate may overlap with only currently stored initial routes and / or associated exclusive safety areas, it may still send a storage permission message to the external control center and / or one or more autonomous vehicles (contrary to the embodiments described above), and instead delete the already stored initial routes and associated safety areas that overlap with the safety area candidate, while further reassigning the new initial routes and associated exclusive safety areas to the one or more autonomous vehicles to which the deleted initial routes were assigned.

[0094] Here, specifically, this can lead to further improvements in the system's route allocation efficiency in some cases, such as the additional possibility of reallocating already existing travel routes and associated safety areas, and the current route allocation can be individually and dynamically adapted to changes in the environment or newly emerging requirements if necessary (i.e., if, for example, no other safe route candidates can be found for a given autonomous vehicle).

[0095] Furthermore, instead of, or in addition to, the above procedure, if the exclusive area manager determines that there are insufficient given route candidates or associated safe area candidates when performing the above security inspection mechanism, the route manager may also preferably calculate its own alternative route candidates, and instead of calculating the associated alternative safe area candidates and sending the alternative route candidates and alternative safe area candidates to the corresponding external control center and / or one or more autonomous vehicles (the external control center and / or one or more autonomous vehicles may further be configured to automatically store the above alternative routes and safe areas for emergency avoidance). This may efficiently avoid additional waiting time in the system.

[0096] Route selection Based on the route candidate generation mechanism described above implemented in the proposed autonomous vehicle management mechanism, each of the current one or more autonomous vehicles may therefore be assigned one or more emergency routes (route candidates) and associated safety areas (each safety area candidate) stored for each autonomous vehicle, and at the same time, each of the stored emergency routes and safety areas may preferably be adapted to and assigned to specific potential hazards recognized before each change in the environment / each object could become an actual hazard to the corresponding autonomous vehicle.

[0097] As a result, within the present invention, in particular, a corresponding avoidance strategy can be easily and efficiently implemented by simply selecting one of the calculated and stored route candidates and associated safety areas for a corresponding autonomous vehicle facing an actual hazard from a given object at a determined time, and switching the initial route and safety area currently used by the autonomous vehicle to the selected route candidate and associated safety area candidate.

[0098] Accordingly, in a preferred embodiment of the proposed autonomous vehicle management system, a given autonomous vehicle, for example, the first autonomous vehicle described above, may further be configured to apply countermeasures to an imminent hazard by selecting one of the stored route candidates and associated safety area candidates, which are assigned to the corresponding autonomous vehicle and, preferentially, to the corresponding object assigned to the detected actual hazard, and by changing the initial route and safety area with the selected route candidate and safety area candidate (i.e., the autonomous vehicle further moves along the selected route candidate).

[0099] Here, the identification of selectable route candidates and associated safety area candidates may be performed efficiently within the system. In particular, since each of the stored route candidates and associated safety area candidates is already assigned to a given autonomous vehicle and a given / recognized object, the corresponding (first) autonomous vehicle facing an object of the "actual hazard" level may further be configured to simply identify the object assigned to the current actual hazard and read the stored route candidates and associated safety area candidates assigned to the identified object from a predetermined memory area where the respective route and area information is preferentially stored.

[0100] In addition, the selection of each route candidate and associated safety area used for hazard avoidance may preferably be based on a predetermined avoidance state defined and managed internally by the autonomous vehicle management system. For example, the autonomous vehicle management system may include at least a "security priority state" and an "efficiency priority state," thereby allowing the selection of corresponding route candidates and associated safety area candidates to be based on the state currently in the autonomous vehicle management system. As an example, there may be first and second route candidates in the first embodiment of the route candidate generation process described above, and if the autonomous vehicle management system is currently in an efficiency priority state, each (first) autonomous vehicle may be configured to select the route candidate and safety area associated with the highest probability of avoiding the corresponding threat (i.e., the first route candidate and safety area in the first embodiment), and in an efficiency priority state, the autonomous vehicle may select an existing route candidate and safety area corresponding to the most efficient (e.g., fastest) means of still reaching the destination (i.e., the second route candidate and associated safety in the first embodiment).

[0101] Integration of vehicle priority and vehicle-restricted areas. In light of the above-described mechanisms and strategies implemented in the present invention, the proposed autonomous vehicle management system, primarily by enabling a high-speed predictive method for generating countermeasures against potential hazards present within the system, thus provides a general-purpose method for further improving the safety and efficiency of current cooperative systems.

[0102] At the same time, the proposed route and safety area generation mechanisms are freely applicable to any type of cooperative system, independently of the type or task assigned to the autonomous vehicle included in the system. Specifically, the adaptive nature of the initial route and route candidate generation process described above also allows for the inclusion of different environmental properties, if necessary. Therefore, the present invention may be implemented more equally and efficiently in closed machining systems or even in road-based delivery systems that typically include impassable or restricted areas.

[0103] For this reason, another preferred embodiment of the present invention may enable the autonomous vehicle management system to calculate and assign the above-mentioned initial routes, route candidates, and associated safety areas by incorporating different local characteristics, such as permitted and prohibited crossing areas, included in the corresponding environment of the system, into the respective mechanisms used in the present invention.

[0104] To this end, for example, an external control center, one or more autonomous vehicles, and / or one of their implemented elements may be configured to further store and access additional map data, which includes two-dimensional information relating to areas of movement in which one or more autonomous vehicles are moving, the two-dimensional information including at least one or more vehicle-priority areas that define areas of movement in each environment of a system in which one or more autonomous vehicles are permitted and / or presumed to move, and one or more vehicle-restricted areas that may define areas of movement in a system in which one or more autonomous vehicles are generally prohibited from moving, and thus information describing a given system environment including different local characteristics.

[0105] In an additional step, at least the route planner of the external control center may also be configured to similarly access and refer to the map data to calculate and assign at least a given initial route, and to utilize only the locations of the system defined as vehicle priority areas within the map data. Conversely, the exclusive area manager of the external control center may also perform the calculation of each exclusive safety area in the same manner, in particular by equally accessing the existing map data and utilizing only the locations of vehicle priority areas to allocate a given exclusive safety area, thereby enabling each of the initial routes and exclusive safety area allocations created in the present invention to be efficiently assigned to specific predetermined travel areas as needed.

[0106] In addition, the same local restrictions may be equally applied during the calculation and allocation of corresponding route candidates and associated safety area candidates, which are performed after hazard detection, preferably by an external control center and / or one or more autonomous vehicles, preferentially limiting the allocation of the above route candidates and safety area candidates to locations associated with vehicle priority areas in their respective map data. In contrast, in different embodiments, it may also be equally possible that locations corresponding to vehicle no-go areas are still acceptable, preferentially increasing the amount of possible potential emergency routes for a given autonomous vehicle in order to calculate each route candidate and safety area candidate. As a result, within this embodiment, an exclusive area manager implementing a security selection mechanism for excluding certain route candidates and safety areas may still be configured to send a storage permission message to the corresponding external control center and / or one or more autonomous vehicles, for example, even if the locations of a given route candidate and / or associated safety area candidate overlap with existing vehicle no-go areas present in the map data.

[0107] As a result, the proposed autonomous vehicle management system of the claimed invention may offer several advantages compared to management systems commonly used in cooperative environments.

[0108] In addition, the aforementioned management system may enable the development of several autonomous vehicle management method steps that have not yet been provided or disclosed by conventional autonomous vehicle management systems.

[0109] Herein, the above method also preferably comprises an external control center, one or more autonomous vehicles communicably connected to the external control center, and a route planner, wherein the external control center may further comprise at least an exclusive area manager, and the route planner may be defined for dynamically managing the driving routines of multiple autonomous vehicles by an autonomous vehicle management system which may be included in the external control center or one or more vehicles. In addition, the method comprises at least, The process involves a route planner calculating an initial route and assigning it to one or more autonomous vehicles, and an exclusive area manager calculating and storing an exclusive safety area for the initial route. The process involves one or more autonomous vehicles autonomously moving from a predetermined starting point to a destination point based on an assigned initial route, The steps include: detecting changes in the environment around an initial route using an external control center and / or one or more autonomous vehicles, and determining at least one of potential or actual hazards for one or more autonomous vehicles moving along the assigned initial route; If a change in the environment around the assigned initial route is determined to be a potential hazard, the external control center and / or one or more autonomous vehicles calculate, for a first autonomous vehicle among one or more autonomous vehicles, at least one alternative route different from the assigned initial route, and an exclusive safety area candidate associated with at least one alternative route, and store the at least one alternative route and the exclusive safety area candidate. If a change in the environment around the initial route is determined to be an actual hazard, the first autonomous vehicle selects one of the stored route candidates. The first autonomous vehicle sets the selected route candidate as the new initial route and sets the selected exclusive safety area candidate associated with the selected route candidate as the new exclusive safety area. The process may include the step of the first autonomous vehicle continuing to travel along a new initial path.

[0110] Furthermore, the method is also, An external control center and / or one or more autonomous vehicles add the stored exclusive safety area candidates to the area assigned to the exclusive safety area of ​​the first autonomous vehicle, An external control center and / or one or more autonomous vehicles send requests to the exclusive area manager for each calculated route candidate, and store each route candidate and the calculated exclusive safety area candidate. The external control center and / or one or more autonomous vehicles will store at least one route candidate and exclusive safety area candidate only if the external control center and / or one or more autonomous vehicles receive a storage permission message from the exclusive area manager. The exclusive area manager of the external control center stores information about each stored exclusive safe area and each stored exclusive safe area candidate, The external control center may, at a minimum, access the stored information and send a storage permission message to the external control center and / or one or more autonomous vehicles if the requested exclusive safety area candidate does not overlap with any of the stored exclusive safety areas or stored safety area candidates. [Brief explanation of the drawing]

[0111] [Figure 1a] This is a schematic diagram illustrating the allocation of exclusive safety areas in a cooperative system according to a first embodiment of the prior art. [Figure 1b] This is a schematic diagram illustrating the allocation of exclusive safety areas in a cooperative system according to the present invention. [Figure 2a] This is a schematic diagram illustrating an example of an exclusive safety area allocated according to the present invention when a potential hazard is detected. [Figure 2b]This is a schematic diagram illustrating the allocated exclusive safety area according to the present invention during route candidate calculation. [Figure 2c] This is a schematic diagram illustrating the allocated exclusive safety areas and allocated exclusive safety area candidates according to the present invention when selecting a route candidate. [Figure 3] This figure illustrates the composition of elements included in the autonomous vehicle and control center of the present invention according to the first embodiment. [Figure 4] This is a flowchart showing the general operating principles of an autonomous vehicle management system according to a predefined embodiment of the present invention. [Figure 5a] This diagram illustrates the workflow for determining the risk level according to the first embodiment. [Figure 5b] This diagram illustrates the workflow for determining the risk level according to the second embodiment. [Figure 6] This diagram exemplifies a workflow in which one or more autonomous vehicles in the system move along an initial path and dynamically allocate a safety area. [Figure 7] This diagram illustrates the workflow for calculating safe path candidates using one embodiment of the present invention. [Figure 8a] This diagram illustrates a workflow for predicting the movement of potential hazards. [Figure 8b] This figure illustrates the predicted movement of Figure 8a, using a person and another vehicle as exemplary hazardous objects. [Figure 9a] This figure illustrates a workflow for calculating first and second path candidates according to one embodiment of the candidate calculation process of the present invention. [Figure 9b] This figure illustrates the candidate routes generated by the workflow shown in Figure 9a. [Figure 10a] This diagram illustrates a workflow for storing and allocating exclusive safety areas according to the first embodiment. [Figure 10b] This diagram illustrates a workflow for storing and allocating exclusive safety areas according to a second embodiment. [Figure 11a] This diagram illustrates the starting position of an autonomous vehicle with an assigned initial route, according to one embodiment of the present invention. [Figure 11b] This diagram exemplifies the system shown in Figure 11a, where the allocated safety area for the autonomous vehicle is visible. [Figure 11c] This diagram exemplifies the system in Figure 11a, where the assigned safety area of ​​the autonomous vehicle is visible, and an object is entering the exclusive safety area. [Figure 11d] This diagram exemplifies the system in Figure 11a, where an autonomous vehicle is moving along an initial path. [Figure 11e] This diagram exemplifies the system in Figure 11d, where the assigned safety area of ​​the autonomous vehicle is visible, and an object is entering the exclusive safety area. [Figure 11f] Figure 11d illustrates the system, showing the assigned safety area for the autonomous vehicle and two objects located near the left and right sides of the autonomous vehicle. [Figure 11g] Figure 11d illustrates the system in which the autonomous vehicle's assigned safety area is visible, the autonomous vehicle is located in a different location from its initial path, and two objects are located near the front and right side of the autonomous vehicle. [Figure 12a] This diagram exemplifies the system of Figure 11a, where the assigned exclusive safety area and route candidates for the autonomous vehicle are visible, the assigned exclusive safety area includes the original area of ​​the exclusive safety area and the calculated exclusive safety area candidate area, and the objects assigned to the route candidates are shown to the left of the autonomous vehicle. [Figure 12b] This diagram exemplifies the system of Figure 11a, where the assigned exclusive safety area and route candidates for the autonomous vehicle are visible, the assigned exclusive safety area includes the original area of ​​the exclusive safety area and the calculated exclusive safety area candidate area, and the objects assigned to the route candidates are shown to the right of the autonomous vehicle. [Figure 12c]This diagram exemplifies the system of Figure 11a, in which the assigned exclusive safety area and route candidates of the autonomous vehicle are visible, the assigned exclusive safety area includes the original area of ​​the exclusive safety area and the calculated exclusive safety area candidate area, and the objects assigned to the route candidates are shown in front of the autonomous vehicle. [Figure 13] This figure illustrates another exemplary embodiment of the composition of the elements in Figure 3, in which a route planner is included in one or more autonomous vehicles. [Figure 14] This diagram exemplifies a concept that includes map data containing vehicle priority areas and vehicle-restricted areas for initial route calculation. [Modes for carrying out the invention]

[0112] Preferred embodiments and models are described in further detail below with reference to the accompanying drawings. The same or similar features in different drawings and embodiments are indicated by similar reference numerals. It should be understood that the following detailed description of various preferred embodiments and models is not intended to limit the scope of the invention.

[0113] Figure 1a shows an illustrative diagram of route and safety area allocation for two autonomous vehicles V1 and V2 using a commonly used autonomous vehicle management system. Here, autonomous vehicles V1 and V2 are each assigned, exemplary, first and second routes IRV1 and IRV2, through which they are required to travel in order to fulfill the transportation requirements assigned to them. Furthermore, both autonomous vehicles V1 and V2 are assigned predetermined exclusive safety areas SAV1 and SAV2, which define areas where the autonomous vehicle is required to temporarily halt its movement if a given object, such as a person or another vehicle, is introduced into the area, thereby enabling collision avoidance when traveling along the assigned route.

[0114] For this reason, in conventional systems such as those shown in Figure 1a, the allocation of exclusive safety areas is typically performed in such a way that potential emergency routes to which autonomous vehicles V1 and V2 must move in the event of an imminent collision crisis are automatically incorporated into the assigned exclusive safety area, thereby enabling the safe movement of the autonomous vehicles. At the same time, even if no emergency action is taken by the autonomous vehicles, a large system area is typically occupied by each of the exclusive safety areas sSAV1 and SAV2. As a result, the area usage required by general autonomous vehicle management systems is still considered inefficient.

[0115] Figure 1b, in contrast, shows the typical exclusive safety area allocation required for exemplary autonomous vehicles V3 and V4 according to the present invention when autonomous vehicles V3 and V4 are moving along their respective initial paths IRV3 and IRV4 and no additional hazards and / or objects are considered to be around autonomous vehicles V3 and V4. As can be seen from the figure, primarily, the autonomous vehicle management system of the present invention is stipulated to perform a dynamic exclusive safety area allocation mechanism that adds additional safety areas as needed, only when necessary (e.g., if a given detected hazard emanating from an object causes one of the autonomous vehicles V3 and V4 to move along a provided emergency path), so that the exclusive safety areas SAV3 and SAV4 for autonomous vehicles V3 and V4 are significantly smaller than the commonly used exclusive safety areas SAV1 and SAV2 shown in Figure 1a. Thus, safety area allocation can be performed much more efficiently, which is equally linked to improved efficiency within a comprehensive cooperative system.

[0116] Figures 2a and 2c further illustrate an overall step-by-step example of the dynamic safety area allocation mechanism introduced by the present invention.

[0117] Here, Figure 2a may use the same autonomous vehicle V3 already shown in Figure 1b, which is currently moving along its initially assigned initial path IRV3 and has its own assigned exclusive safety area SAV3. Furthermore, in the illustrated situation, there is an object O depicted as a person moving next to the autonomous vehicle V3, defining a potential collision risk for the autonomous vehicle V3. As a result, in order to avoid potential collisions while efficiently preventing the autonomous vehicle V3 from pausing its current movement, the autonomous vehicle management system may be configured to dynamically calculate anticipated emergency routes (so-called "route candidates," CR1 and CR2 shown in Figure 2b) before an actual hazardous event is detected (for example, when object O enters the autonomous vehicle V3's exclusive safe area SAV3), and to allocate additional exclusive safe area candidates (CSA1 and CSA2 in Figure 2c) to the autonomous vehicle V3, which are added to the autonomous vehicle V3's exclusive safe area SAV3 and associated with the calculated emergency routes CR1 and CR2, thereby providing a safe area allocation mechanism that is more efficient but equally safe.

[0118] Figure 3 shows elements included in the proposed autonomous vehicle management system of the present invention, as well as more preferred embodiments of their preferred compositions.

[0119] Here, specifically, the proposed autonomous vehicle management system may include at least one or more autonomous vehicles V capable of traversing a predefined area of ​​the system, and in this embodiment, a control center 300 comprising at least a route planner 302 and an exclusive area manager 304. Furthermore, in order to move one or more autonomous vehicles V, each autonomous vehicle V requires a move instruction contained in the initial route (exemplified herein by element 404), which is received by the autonomous vehicles V, preferably primarily by using a predefined communication connection, by first sending a route request 402 to the route planner 302 of the external control center, and then receiving an initial route 404 assigned to each autonomous vehicle V by the route planner 302, in the embodiment of Figure 3. The initial route 404 itself may further include all necessary instructions, such as time, speed, start location, end location, and the method by which the autonomous vehicles V are instructed to move, and the autonomous vehicles V are further automatically configured to execute the instructions contained in the initial route.

[0120] In addition, with respect to the dynamic adjustment of the initial route, including dynamic emergency (i.e., candidate) route calculation, allocation of additional exclusive safety area candidates, and switching to the emergency route if necessary, one or more autonomous vehicles may each further include a local route calculation feedback loop represented by element 406 in Figure 3.

[0121] In this case, the local route calculation feedback loop 406 may include several functions and / or mechanisms implemented in this embodiment of Figure 3 by each autonomous vehicle V itself (or by one or more modules included in each autonomous vehicle V), and in other embodiments, such functions and mechanisms may be equally performed by other elements of the proposed autonomous vehicle management system, such as an external control center 300, or may be performed in parallel and partially by the vehicle V and the external control center 300. For this reason, when the initial route 404 is received from the control center 300, the local route calculation feedback loop 406 may first begin by reading the exclusive safe area associated with the initial route 404. To this end, the autonomous vehicle V may send an exclusive area request 408 to an exclusive area manager 304 that is communicably connected to the autonomous vehicle V, and the exclusive area manager 304 may receive the exclusive safe area after calculating, storing, and assigning the associated exclusive safe area to each autonomous vehicle V.

[0122] Simultaneously, in order to permanently consider dangerous encounters while traveling along the initial path 404, the autonomous vehicle V continuously calculates, stores, and updates new emergency paths, so-called safe path candidates and associated exclusive safe area candidates, using a permanent feedback loop of the local path calculation feedback loop 406, depicted by elements 408, 410, 412, and 416 in Figure 3. Specifically, the feedback loops include, at a minimum, a series of actions: recognizing environmental changes, i.e., objects present around the initial path 404 of the autonomous vehicle V (defined by the environmental recognition element 414), and subdividing the recognized objects into different hazard levels that describe the currently anticipated hazard of the objects. Furthermore, depending on the results of the detection, each feedback loop further continues either adaptively calculating additional safe path candidates and additional associated exclusive safe area candidates, or selecting a given safe path candidate and safe area for collision avoidance.

[0123] More specifically, if the hazard of a given recognized object is determined to be a potential hazard (defined by the “Potential Hazard Detection” element 412 in Figure 3), that is, an object that may become an actual hazard within a given time period, the autonomous vehicle V may perform an additional route candidate calculation step, in which an emergency route to avoid the object (if it becomes an actual hazard at some point) is calculated and stored with reference to the respective autonomous vehicle V, represented by the “Safe Route Candidate Calculation” element 410 in Figure 3. In addition, an additional exclusive safe area candidate associated with the calculated route candidate is also generated and preferably allocated to the autonomous vehicle V (again, by requesting a calculation from the exclusive area manager 304 of the external control center 300 to calculate an appropriate exclusive safe area candidate using the exclusive area request 408), thereby enabling the dynamic and adaptive generation of safe areas for each autonomous vehicle V in the system.

[0124] In contrast, if a given object is an actual hazard (hereinafter also defined by the hazard level of the “actual hazard” as defined by hazard detection 416 in Figure 3), that is, if the object clearly defines a threat to the autonomous vehicle V, the autonomous vehicle V may perform adaptive route selection 420, thereby performing collision avoidance by selecting one of the previously stored and assigned route candidates and associated exclusive safety area candidates, and then switching its dynamic route movement to instructions contained in the selected route candidate (defined by the route follower 422, i.e., the autonomous vehicle V follows the route candidate instead of the initial route).

[0125] Therefore, this enables the provision of continuous, adaptive, and efficient route and safety area allocation strategies.

[0126] Figure 4 illustrates the workflow described above for the embodiment of the present invention shown in Figure 3, this time illustrating it with a step-by-step flow sheet. Here, the left side of the figure may define the processes performed by the external control center 300, and the right side defines the processes performed by at least one of the one or more autonomous vehicles V of the system. Again, the calculation of route candidates and associated exclusive safety area candidates is handled here by one or more autonomous vehicles V, but similarly, and preferably in other embodiments, it can be performed by other elements of the system.

[0127] Therefore, in step SB01, in order to start the autonomous vehicle V moving, the autonomous vehicle V may first request to receive instructions from the route planner 302 of the external control center 300 regarding an initial route for moving from starting point A to destination point B.

[0128] In response, the route planner 302 of the external control center 300 receives the request from the autonomous vehicle V, calculates an appropriate initial route in step SA01, and sends the initial route, including instructions, back to the autonomous vehicle V.

[0129] In step SB02, the autonomous vehicle V further requests the exclusive safe area associated with the received initial route from the exclusive area manager 304.

[0130] In response, in step SA02, the exclusive area manager 304 receives the request, calculates the associated exclusive safety area, and then sends it back to the autonomous vehicle V. The autonomous vehicle V then receives the associated exclusive safety area and, referring to the autonomous vehicle V, stores and / or allocates both the initial route and the exclusive safety area. During this time, preferably, the autonomous vehicle V may be prohibited from moving until the reception, storage, and / or allocation of the initial route and the exclusive safety area has been sufficiently performed.

[0131] Subsequently, the autonomous vehicle V may move along the assigned initial path. During this time, the calculation of the adaptive path candidates and exclusive safety area candidates, as well as the hazard avoidance mechanism described above, may be continuously performed, including the following steps.

[0132] In step SB03, environmental awareness is performed by the autonomous vehicle V. That is, the autonomous vehicle V detects current changes in the environment, for example, detecting obstacles present along the initial path.

[0133] Simultaneously, as shown in step SB04, the autonomous vehicle V may continuously access the instructions included in the initial path and make adaptations to the vehicle's current characteristics (e.g., changing the vehicle V's speed, direction, etc.) to follow the initial path.

[0134] Furthermore, in step SB05, based on the environmental perception performed in step SB03, it is determined whether a potential hazard may exist in the vicinity of the autonomous vehicle V. If such a potential hazard exists, the autonomous vehicle V performs a safe route candidate calculation SB06, which includes calculating, assigning, storing, and potentially allocating one or more route candidates and associated exclusive safe area candidates to avoid a potential collision caused by the potential hazard.

[0135] Furthermore, if no potential hazards are found, the autonomous vehicle V will assess in step SB07 that no real, or actual, hazards were found.

[0136] If no actual danger is found, the autonomous vehicle V continues to move along the initial path in step SB08, repeating the detection steps described above until it arrives at a designated location.

[0137] In contrast, if an actual hazard is identified in step SB07, the autonomous vehicle further assesses whether appropriate safe route candidates are stored and assigned to each hazardous object.

[0138] If no suitable route candidates are stored, the autonomous vehicle V will apply full brakes as a last resort, thus avoiding a given collision by completely halting its movement (step SB09).

[0139] Alternatively, if suitable route candidates are stored, it may be detected whether each route candidate can be fully continued or whether another threat may be present in the vicinity of the selectable stored route candidates, and if it is determined that the stored route candidates can be continuously followed (step SB11), the autonomous vehicle V may select a route candidate and follow it until it reaches the destination, or the autonomous vehicle may select and follow each route candidate and then stop at a predetermined location.

[0140] Figure 5a further illustrates an exemplary embodiment of the potential hazard detection mechanism shown in step SB05.

[0141] Specifically, in this embodiment, potential hazards may be determined by determining the distance from a given recognized object O to an autonomous vehicle V, and defining object O as a potential hazard if the distance is smaller than a predetermined potential hazard threshold.

[0142] Therefore, for this purpose, the autonomous vehicle V may, in the first step SB071a, receive information about each object recognized during the environmental recognition SB03, and this information may include at least the location of each object. For this purpose, for example, the autonomous vehicle V may access a predetermined memory area in which information about each recognized object is continuously stored within the system of the present invention.

[0143] Subsequently, the autonomous vehicle V may equally access its own position information (SB072a) and then calculate the distance from the autonomous vehicle V to object O by determining whether the calculated distance is smaller than a predetermined potential danger threshold (SB073a). Therefore, if the distance is indeed small, the autonomous vehicle V may further assign and register the object as a potential danger and follow the steps of SB06 shown in Figure 4 (SB074a).

[0144] Furthermore, the same method may be applied to detect actual hazards within the system, particularly by using an additional actual hazard threshold smaller than the potential hazard value, and determining actual hazard when the distance is perceived to be smaller than the actual hazard threshold (in this case, determining potential hazard would require a distance smaller than the potential hazard threshold but larger than the actual hazard threshold).

[0145] For this reason, the right side of Figure 5a may show a given example in which each of the hazard levels may exist after performing the hazard determination mechanism in Figure 5a. Here, the actual hazard threshold may be indicated by line 502, and the potential hazard threshold may be defined by line 503. As a result, object O3 will be considered an actual hazard, object O2 will be considered a potential hazard, and object O1 will be considered no hazard at all.

[0146] In addition, Figure 5b shows another embodiment of the hazard detection step of SB07, where instead of only the distance between the vehicle V and the object O, the position and velocity of each object are received as parameters (SB071b), and the hazard level of a given object O is determined by predicting the distance from object O to vehicle V at a predetermined time in the future. Specifically, in this case, the autonomous vehicle V may calculate the trajectory, or in this case the x and y components of the two-dimensional trajectory of each object O, based on the received position and velocity information (SB072b), and determine, according to the detection mechanism described above, whether the distance from vehicle V to the trajectory at a given time is less than predetermined thresholds, such as the potential hazard threshold and the actual hazard threshold described above (SB073b, SB074b). In this way, a more accurate assessment of the hazard level of object O can be made possible.

[0147] Here, to the right of Figure 5b, an additional diagram of the determination mechanism of Figure 5b is shown, in which the first object O4 may show a first trajectory VE4 that always provides a sufficient distance to the vehicle V, and the second trajectory VE5 of the second object O5 may show an insufficient distance at a certain point in time. Therefore, O5 may be determined to be a potential hazard, and O4 may be considered not to be a hazard.

[0148] Figure 6 further provides a detailed description of the continuous adaptive and mobile mechanisms performed by the autonomous vehicle V during movement, as described by step SB04 in Figure 4, or by SB08, respectively.

[0149] Here, the autonomous vehicle V shown in this embodiment may be equipped with a dynamic movement mechanism that requires the presence of an appropriate exclusive safety area whenever the next movement is performed. Specifically, the movement mechanism shown in Figure 6 is defined such that the autonomous vehicle V may move only in directions where an exclusive safety area already exists, and in addition, further exclusive safety areas may be continuously allocated and assigned to the autonomous vehicle V as needed to enable the vehicle's continued safety.

[0150] Therefore, based on the embodiments described above, the movement mechanism SB04 / SB08 of the autonomous vehicle V may be initiated by a request to move further along the initial path assigned to the autonomous vehicle V.

[0151] To this end, the autonomous vehicle V may, in the first step SB081, detect whether the next movement step to be performed to follow the initial path is already ensured by the assigned exclusive safety area, that is, whether the next movement step is still within the area defined by the currently assigned exclusive safety area.

[0152] If it is determined that the next movement step is still certain by the currently allocated exclusive safety area, the autonomous vehicle V may further check whether a new preferred exclusive safety area can be safely allocated to the autonomous vehicle V in the next movement step as well, so as to enable the subsequent movement step (SB042).

[0153] If both of the above criteria are currently met with respect to the autonomous vehicle V, the autonomous vehicle V may proceed to the next appropriate movement step indicated by the initial path and may further retrieve the initial path in step SB045. In addition, an additional exclusive safety area allocation process may be performed in step SB088 by requesting and retrieving updated exclusive safety areas from the exclusive safety area manager 304 of the external control center 300 to provide a sufficient basis for the next movement step of the autonomous vehicle V.

[0154] Conversely, if autonomous vehicle V cannot proceed to the next move step because it is not currently assigned an exclusive safety area where the next move step would be ensured (i.e., the next move step would involve entering an area not included in the exclusive safety area currently assigned to autonomous vehicle V), autonomous vehicle V may check whether the exclusive safety area necessary to proceed further is currently assigned to / allocated to another vehicle's current exclusive safety area candidate (SB083).

[0155] If this is not the case (i.e., the required area is instead being used by another vehicle located in an exclusive safety area instead of a candidate safety area), the autonomous vehicle may instead wait for a predetermined amount of time (SB084 / SB085) and then try to request the required area again.

[0156] Alternatively, if the required area is found to be part of another vehicle's exclusive safety area candidate, the autonomous vehicle V may rather send a request to the route planner 302 of the external control center 300 to receive an alternative route to reach the designated destination (SB089), and then proceed along the alternative route after receiving it by equally requesting a contiguous external safety area allocation from the exclusive area manager 304.

[0157] Figure 7 further shows a first preferred embodiment for calculating route candidates, including those included in step SB06 shown in Figure 4.

[0158] Specifically, the path candidate calculation mechanism depicted in Figure 7 is based on receiving spatial information of objects associated with corresponding potential hazards, as well as using a prediction algorithm during movement.

[0159] To this end, in the first step SB061, the autonomous vehicle may receive information, specifically spatial information such as the location, speed, and / or trajectory of the corresponding object.

[0160] Subsequently, the autonomous vehicle V may predict the movement of objects and predict the future positions of one or more objects by analyzing the retrieved information (SB062).

[0161] Next, the autonomous vehicle V may begin calculating appropriate route candidates by first detecting the position of the object at the estimated time. Specifically, the autonomous vehicle may first check whether the object is located in front of the vehicle V (step SB063), to the right of the vehicle V (SB065), or to the left of the vehicle V at the estimated time, and then appropriately calculate one or more appropriate route candidates (steps SB064, SB066, SB067).

[0162] Furthermore, after sufficiently calculating one or more route candidates, the autonomous vehicle V may further require an exclusive safety area candidate associated with each of the calculated route candidates. To this end, the autonomous vehicle V requests the corresponding exclusive safety area candidates from the exclusive area manager 304 of the external control center 300, in particular by sending a candidate area calculation request to the exclusive area manager 304 (step SB068). The exclusive area manager 304 then calculates each exclusive safety area candidate, stores them if possible, assigns and allocates them to the associated route candidates (SA05), and finally returns them to the autonomous vehicle V. Furthermore, in possible embodiments, the exclusive area manager 304 may also perform an additional security selection process at this point, thereby omitting even successfully calculated exclusive safety area candidates; that is, if a calculated exclusive safety area candidate does not meet predefined requirements set by the system, it may not be returned to the autonomous vehicle V.

[0163] Next, the autonomous vehicle V may examine the exclusive safety area candidates received from the exclusive safety area (SB069), and then store such route candidates for which the relevant exclusive safety area candidates have been received (SB0610). In other words, only such calculated route candidates are stored and can be used by the autonomous vehicle V, encompassing at least one relevant exclusive safety area candidate in the system, and as a result, the exclusive safety area candidate calculation process performed by the exclusive area manager 304 can also be viewed as an efficient pre-selection process for each of the calculated route candidates.

[0164] For this reason, in different embodiments, it is equally possible that the calculation of associated exclusive safety area candidates is not performed by the exclusive area manager 304 itself, but instead by one of the other elements present in the autonomous vehicle management system. In these embodiments, the exclusive area manager 304 may still be configured to perform pre-selection of calculated route candidates by sending a store permission request to the exclusive area manager 304, which includes, more preferably, preferentially, the calculated route candidates and, where applicable, information about already calculated exclusive safety area candidates assigned to the route candidates, and then storing the calculated route candidates (and associated exclusive safety area candidates) only if the exclusive area manager 304 resends a store permission message to the autonomous vehicle V for each of the route candidates after performing an additional selection mechanism (e.g., a security selection mechanism).

[0165] Figure 8a further shows a more detailed diagram of one embodiment of the object movement prediction strategy (step SB062) used in the embodiment of Figure 7, which predicts the location of potentially hazardous objects for path candidate calculation.

[0166] Here, each strategy may first include recognizing the type of each object O in the first step (SB0621). For this purpose, for example, the autonomous vehicle V may be configured to store specific / detected characteristics of the recognized object O, such as size, shape, and / or speed, in the system's memory and compare them with predetermined characteristics assigned to defined object types (e.g., human, car, motorcycle, bicycle, etc.) to thereby identify each type of recognized object O.

[0167] Subsequently, depending on the results of the object type identification described above, different movement prediction strategies may be implemented by the autonomous vehicle V.

[0168] For example, if an autonomous vehicle V can recognize the object type of a corresponding object O as an object having wheels and handlebars (e.g., a car, motorcycle, or bicycle), the autonomous vehicle V may further be configured to predict the object's movement by preferentially utilizing a non-holonomic system model explicitly assigned to the object type of each object (SB0622). Based on this, the autonomous vehicle may predict, for example, the vector length and direction of the trajectory potentially taken by the object based on the respective model and received information, and then compute one or more path candidates based on the thus generated trajectory vector.

[0169] In addition, in the embodiment shown in Figure 8a, the autonomous vehicle V not only outputs a predicted movement of object O using a prediction of the object O's sustained movement behavior, but also outputs additional vectors that define the variability of the object's movement (e.g., left or right turn in a given amount of time, step SB0623, etc.). As a result, the efficiency of the corresponding path candidates can be further improved, in particular, because each possible movement of object O can be included in subsequent calculations of the path candidates.

[0170] In contrast, in step SB062, the autonomous vehicle V may be able to identify the object type of the corresponding object O as a freely moving object (e.g., a person), or it may not be able to identify the object type of the corresponding object O, and the movement prediction strategy in Figure 8a may be specified to calculate a prediction vector based on a more general policy, such as by using a linear interpolation model (SB0624), and to output not only one prediction vector, but also several variations, such as vectors representing a 45-degree turn of the object in each direction from the original predicted movement vector (SB0625).

[0171] Furthermore, Figure 8b also illustrates a different motion prediction strategy, again drawn according to Figure 8a. Here, the upper part of Figure 8b shows a case where the object type of object OP cannot be sufficiently identified by the autonomous vehicle V (for example, because the object is related to a person who is difficult to identify). In this case, the motion prediction method may calculate a prediction vector VEOP2 based on a linear interpolation model, and then add VEOP1 and VEOP3 to calculate suitable path candidates.

[0172] In contrast, the lower part of Figure 8b shows the case where object OV is sufficiently identified as a vehicle, specifically as a forklift, thereby the motion prediction strategy calculates the prediction vector VEOV2 and variability VEOV1 and VEOV3 according to a non-holonomic system model specifically assigned to the identified object type.

[0173] Figure 9a further shows an additional embodiment of the route candidate calculation mechanism performed in step SB06 of Figure 4.

[0174] Specifically, the embodiment in Figure 9a includes a mechanism for calculating at least two different path candidates for each potential hazard detected in the previous hazard detection step SB05, and simultaneously, to calculate these path candidates, the prediction information obtained from the movement prediction mechanism in Figure 8 is used to calculate both path candidates.

[0175] For this reason, in the first step SB06A1, an iterative loop is initiated in which each of the following process steps is executed iteratively for each predicted movement vector previously generated by the movement prediction mechanism.

[0176] Therefore, in the second step SB06A2, each predicted movement vector is iteratively selected and then used in the next step SB06A3 to calculate a first path candidate. Here, the first path candidate may be calculated, for example, by using a predetermined calculation algorithm, such that the maximum distance from the autonomous vehicle V to object O is always maintained, resulting in a safety-prioritized path candidate.

[0177] Next, after sufficiently calculating the first path candidate for each predicted movement vector, the first path candidate is registered and stored (SB06A4), and the calculation of the second path candidate (SB06A5) begins.

[0178] Here, in contrast to the first route candidate, the second route candidate may be calculated, specifically, based on a second predetermined calculation algorithm, such that a specific distance from the autonomous vehicle V to object O is always maintained, and at the same time, it is possible to connect the second route candidate to the initial route at some point (thus defining the second route candidate as a short route prioritizing transport efficiency).

[0179] Finally, after calculating the second route candidate, the latter is equally registered / stored by the autonomous vehicle V (SB06A7), and the preceding step is repeated again for another predicted movement vector (or the calculation ends in step SB068, respectively).

[0180] Figure 9b further shows the characteristics of the first and second path candidates, respectively, calculated by the candidate calculation mechanism in Figure 9a. Specifically, the upper part of Figure 9b shows the orientation of the first path candidate CR1 used by the autonomous vehicle V, which is oriented away from the predicted movement vector VEO of object O, and therefore allows for the maximum distance to the object. Conversely, the lower part of Figure 9b shows that the second path candidate CR2 remains rather close to the initial path of vehicle V, and even reconnects to it after a certain amount of time, resulting in an efficient collision avoidance process.

[0181] Figure 10a further illustrates an exemplary workflow of one embodiment of the exclusive area selection and allocation mechanism used in the present invention, which is performed by the exclusive area manager 304 to calculate and select sufficient safe areas to be assigned to the relevant initial route or route candidate. Thus, each process step may be used to calculate both the exclusive safe area (SA04) assigned to the initial route of the autonomous vehicle V, or the exclusive safe area candidate (SA05) assigned to the route candidate. At the same time, the same process may also be used in or included in the security selection process of the exclusive area manager 304 described above, which is not yet explicitly shown herein.

[0182] For this reason, in the first step SA041 of the exclusive safety area selection mechanism in Figure 10a, the exclusive area manager 304 may receive a requested exclusive safety area and / or exclusive safety area candidate that has been previously calculated by the exclusive area manager 304, the autonomous vehicle V, or any other element of the autonomous vehicle management system, which will be considered for allocation (i.e., assignment to a given initial route or route candidate).

[0183] In the next step, SA042, the exclusive area manager 304 may receive information about each of the currently stored safe areas, i.e., primarily the location areas of each of the exclusive safe areas and exclusive safe area candidates in the system that are currently allocated or in use. To this end, the exclusive area manager 304 may store the above information in a format of its own choosing, for example, by storing each piece of information in a predetermined memory area, or at the very least, it may read the information from a memory area in which each piece of information is already stored. As a result, the exclusive area manager gains access to each area in the system that is currently allocated or assigned to an exclusive safe area or exclusive safe area candidate, which may all be grouped together into a single exclusive safe area map.

[0184] Subsequently, in order to test the sufficiency of each requested safety area, the exclusive area manager 304 may perform a test to determine whether at least the requested safety area or at least a portion of the requested safety area is currently in use, i.e., allocated within the system (SA043). To do this, the exclusive area manager 304 may read the locations of the areas defined by the requested safety area from the received safety area information and compare the locations of each area with the areas in the system to which exclusive safety areas or candidate exclusive safety areas are currently allocated (SA044).

[0185] Next, if the exclusive area manager 304 identifies that the location of the area defined by the requested safety area is not currently allocated or in use by any of the autonomous vehicles V, the exclusive area manager 304 may allocate the requested area to each assigned initial route or route candidate (and therefore to each autonomous vehicle V associated with the route, step SA045), and may further return the approved and allocated safety area to the corresponding autonomous vehicle (SA047).

[0186] In addition, the exclusive area manager 304 may also update the currently stored information about the safe areas by similarly storing the newly allocated safe areas in the corresponding memory area (SA048) and then ending the safe area selection.

[0187] In contrast, if the requested safety area is already stored / used by one of the corresponding autonomous vehicles V, the exclusive area manager 304 may, instead of allocating and storing the requested safety area, send a failure message to each autonomous vehicle V requesting storage and / or allocation of the safety area (or conversely, not send a storage approval message) (SA046), and as a result, the calculated initial route or route candidate is usually omitted or updated (see Figure 4 again).

[0188] Furthermore, Figure 10b illustrates another embodiment of the safety area selection mechanism described above, and the difference between the mechanisms in Figure 10a and Figure 10b is defined by an alternative process performed by the exclusive area manager 304 when it is identified that the location area of ​​the requested safety area is already in use by another exclusive safety area or candidate exclusive safety area in the system.

[0189] In this embodiment, contrary to the case where each requested safety area is not allocated and stored, the exclusive area manager 304 may instead further identify whether the requested safety area belongs to an exclusive safety area candidate (SA046'), and then, if each stored safety area currently blocking the area required by the requested safety area is allocated only to the initial route of another autonomous vehicle V (SA047'), and no other objects or vehicles are currently present in the requested area (SA048'), the allocation and storage of the requested safety area may still be approved (SA0410' and SA048). Thus, safety in the system may be further improved, in particular, by performing these additional steps, as the allocation of exclusive safety area candidates, and therefore the generation and presence of route candidates, may be prioritized within the system. In addition, to compensate for the requested allocation of safety areas, the exclusive area manager 304 may further erase previously stored safety areas (SA049') and preferably reassign new initial routes and exclusive safety areas to each autonomous vehicle V to which the erased safety areas were previously assigned.

[0190] Figure 11a shows yet another diagram of an embodiment of the cooperative system according to the present invention, in order to further demonstrate the mechanisms implemented in each autonomous vehicle management system, in which an autonomous vehicle V is shown at a given starting point A and is assigned an initial route IR to which the autonomous vehicle V must travel to a destination location B.

[0191] In addition, to further illustrate the preferred movement behavior of the autonomous vehicle V according to the present invention and the importance of the exclusive safety area assigned to a given autonomous vehicle V, Figures 11b and 11c show a situation in which the autonomous vehicle V cannot move along the initial path IR based on the assigned exclusive safety area SA. Specifically, in this case, it has been found that an object O enters the vehicle V's assigned exclusive safety area SA and temporarily stops any movement of the autonomous vehicle V (refer again to the movement mechanism shown in Figure 6).

[0192] Figure 11d further shows that each autonomous vehicle V is allowed to move along the initial path to some extent, resulting in a different location for each autonomous vehicle V compared to Figures 11a-11c, and simultaneously, in particular, that the exclusive safety area SA assigned to the autonomous vehicle V is adapted to the movement by repeatedly assigning new location areas to the assigned exclusive safety area SA. In comparison, the same movement characteristics of the autonomous vehicle V shown in Figures 11b and 11c still apply to later stages of movement, and therefore, if object O enters the (newly) assigned exclusive safety area of ​​vehicle V in Figure 11e again, it causes the autonomous vehicle V to pause its movement again.

[0193] Figures 11f and 11g show, in contrast, illustrative diagrams of hazard detection mechanisms used in the present invention. Two objects O1 and O2 shown in Figure 11f, and O3 and O4 shown in Figure 11g, are not in the exclusive safety area SA of the vehicle V, and the autonomous vehicle V may move freely along the initial path IR. At the same time, since objects O1 to O4 are each already in the vicinity of the autonomous vehicle V, the autonomous vehicle may initiate its hazard detection mechanism by first recognizing the presence of the objects, and preferably specific information about each object in the system. As a result, in the systems shown in Figures 11f and 11g, the autonomous vehicle V may, for example, recognize an object on the left side of the vehicle (object O1) and an object on the right side (object O2) in the system of Figure 11f, and an object on the right side of the vehicle (object O3) and an object on the front side (object O4) in the system of Figure 11g.

[0194] Figures 12a, 12b, and 12c further illustrate different scenarios in which the autonomous vehicle V is already using hazard detection, path candidate, and exclusive safe area candidate calculation mechanisms implemented within the system to avoid collisions with moving objects O that are preferentially detected as potential hazards (Figures 12a, 12b, and 12c may differ only by the position and movement of object O, by reference).

[0195] Here, in each of the situations, each autonomous vehicle V can calculate multiple potential emergency routes, specifically first and second route candidates CR1 and CR2 which are drawn preferentially with reference to Figures 9a and 9b in this case, and similarly, there is enough space to allocate the respective exclusive safety area candidates CSA1 and CSA2 assigned to the first and second route candidates. Thus, by selecting one or each of the existing route candidates and moving along it, the autonomous vehicle V is provided with multiple efficient collision avoidance possibilities, while simultaneously minimizing the number and size of the required safety area allocation each time.

[0196] Figure 13 further illustrates a second further embodiment of the proposed configuration of the autonomous vehicle management system according to the present invention. In this embodiment, the elements and mechanisms included may be the same as those in the configuration already described with reference to Figure 3. In contrast, the only difference from the embodiment in Figure 3 may be indicated by the location of the route planner 302', in which case it is located on one or more autonomous vehicles V instead of the external control center 300'. Thus, mainly this makes it possible to directly connect the route planner 302' and route calculation feedback loop of each autonomous vehicle V, and as a result this leads to more stable and error-free signal conversion between each of the existing elements, so it may be possible to further improve the efficiency of the management system in this configuration.

[0197] Figure 14 illustrates yet another concept that should be included in the proposed autonomous vehicle management system. Specifically, since certain cooperative systems (for example, in a road-based system that includes both vehicle movement areas and pedestrian walking areas) may require that an autonomous vehicle V not move along a given area, another possible mechanism to be implemented in the initial path calculation and exclusive safety area calculation mechanism described above may provide vehicle prohibition and vehicle priority areas. Here, a given system may define specific areas within the system environment that one or more autonomous vehicles may access for movement (vehicle priority areas, see, e.g., PA1, PA2, PA3, PA4 in Figure 14), and areas where movement of one or more autonomous vehicles is prohibited (vehicle prohibition areas, see, e.g., VPA1, VPA2, VPA3 in Figure 14), even before the initial path is calculated.

[0198] To include these areas in the initial route calculation, and preferably in the route candidate calculation and associated safety area calculation, elements of the proposed autonomous vehicle management system may further be configured to receive information about the location of each of the predetermined priority and prohibited areas within the system, by prohibiting route and safety area calculations along at least vehicle prohibited areas, and to adapt the vehicle route and safety area calculations according to these predetermined areas (see, for example, initial route IR2, which is prohibited from crossing a vehicle prohibited area, specifically vehicle prohibited area VPA2, while initial route IR1 is still in use). In this way, specific local characteristics within the environment can be easily incorporated into the mechanism of the proposed invention, while still easily maintaining the efficiency and safety of each management system.

[0199] Furthermore, it should be noted that embodiments of this disclosure may take the form of hardware embodiments as a whole, software embodiments (including firmware, resident software, microcode, etc.), or embodiments combining software and hardware aspects. Moreover, embodiments of this disclosure may take the form of computer program products on computer-readable media in which computer executable program code is embodied on the medium.

[0200] It should be noted that arrows may be used in drawings to represent communication, transfer, or other operations involving two or more entities. A double-headed arrow generally indicates that an operation can occur in both directions (e.g., a command / request in one direction and a corresponding reply in the other, or peer-to-peer communication initiated by either entity), although in some situations the operation does not necessarily have to occur in both directions.

[0201] While a one-sided arrow generally indicates exclusive or primarily unidirectional work, it should be noted that in certain situations, such directional work may actually involve bidirectional work (e.g., a message from sender to receiver and a reply of receipt from receiver to sender, or the establishment of a connection before forwarding and the termination of a connection after forwarding). Therefore, the type of arrow used in a particular diagram to represent a particular task should be considered illustrative and not limiting.

[0202] Examples and embodiments are described herein with reference to flowcharts and / or block diagrams of methods and apparatus. It will be understood that each block in a flowchart or block diagram, or both, and combinations of blocks in a flowchart or block diagram, or both, can be implemented by computer executable program code.

[0203] Any computer-executable program code may be provided to the processor of a general-purpose computer, a dedicated computer, or other programmable data processing device to create a particular machine, thereby creating means for the program code to be executed via the computer or other programmable data processing device to realize functions / actions / outputs specified in flowcharts, blocks of block diagrams, drawings, and / or descriptions.

[0204] These computer-executable program codes may also be stored in computer-readable memory, which can instruct a computer or other programmable data processing device to function in a particular manner, thereby creating a product in which the program code stored in computer-readable memory includes instruction means to realize a function / action / output specified in a flowchart, block diagram blocks, drawing, and / or description.

[0205] Computer-executable program code may also be loaded into a computer or other programmable data processing device to perform a series of operational steps on the computer or other programmable device, thereby generating a computer implementation process, in which the program code executed on the computer or other programmable device provides steps to realize a function / action / output specified in a flowchart, block diagram block, drawing, and / or description. Alternatively, the steps or actions realized by the computer program may be combined with steps or actions realized by an operator or person to carry out the embodiment.

[0206] Communication networks may generally include public and / or private networks, and may include local area, wide area, metropolitan area, storage, and / or other types of networks, and may employ communication technologies in any form, including but not limited to analog technology, digital technology, optical technology, wireless technology (e.g., Bluetooth®), networking technology, and internetworking technology.

[0207] It should also be noted that the device may use communication protocols and messages (e.g., messages created, sent, received, stored, and / or processed by the device), and such messages may be carried by a communication network or medium.

[0208] Unless otherwise required in the context, this disclosure should not be construed as being limited to any particular type of communication message, communication message format, or communication protocol. Therefore, communication messages may generally include, but are not limited to, frames, packets, datagrams, user datagrams, cells, or other types of communication messages.

[0209] Unless otherwise necessary in the context, references to specific communication protocols are illustrative, and it should be understood that alternative embodiments may, where appropriate, be variations of such communication protocols (e.g., modifications or extensions to protocols, which are often done), or other protocols that are known or may be developed in the future.

[0210] Furthermore, it should be noted that while logical flows may be described herein to demonstrate various aspects, this disclosure should not be construed as limiting it to any particular logical flow or implementation. The logic described may be divided into different logical blocks (e.g., programs, modules, functions, or subroutines) without altering the overall result.

[0211] In many cases, logic elements may be added, modified, omitted, implemented in a different order, or implemented using different logic constructs (e.g., logic gates, loop-type primitives, conditional logic, and other logic constructs) without changing the overall result.

[0212] This disclosure may be embodied in many different forms, including, but not limited to, computer program logic used with a processor (e.g., a microprocessor, microcontroller, digital signal processor, or general-purpose computer), programmable logic used with a programmable logic device (e.g., a field-programmable gate array (FPGA) or other PLD), discrete components, integrated circuits (e.g., application-specific integrated circuits (ASICs)), or any other means including any combination thereof. Computer program logic that implements some or all of the described functionality is generally implemented as a set of computer program instructions that are translated into a computer-executable form, stored on a computer-readable medium, and executed by a microprocessor under the control of an operating system. Hardware-based logic that implements some or all of the described functionality may be implemented using one or more appropriately configured FPGAs.

[0213] Computer program logic that implements all or part of the functionalities described above may be embodied in various forms, including but not limited to source code, computer executable forms, and various intermediate forms (e.g., forms generated by an assembler, compiler, linker, or locator).

[0214] The source code may include a set of computer program instructions implemented in one of several programming languages ​​(e.g., object code, assembly language, or high-level languages ​​such as Fortran, C, C++, Java, Python, or HTML) for use with various operating systems or operating environments. The source code may define and use various data structures and communication messages. The source code may be in a form executable by a computer (e.g., via an interpreter), or the source code may be converted into a form executable by a computer (e.g., via a translator, assembler, or compiler).

[0215] Computer executable program code for performing the operations of the embodiments of this disclosure may be written in an object-oriented, scripting, or non-scripting programming language, such as Java®, Perl, Smalltalk, or C++. However, computer executable program code for performing the operations of the embodiments may also be written in a conventional procedural programming language, such as the C programming language or a similar programming language.

[0216] The computer program logic that implements all or part of the functionality described herein may be executed on a single processor at different times (e.g., simultaneously), or on multiple processors at the same or different times, and may run under a single operating system process / thread or under different operating system processes / threads.

[0217] Therefore, the term “computer process” may generally refer to the execution of a set of computer program instructions, regardless of whether different computer processes run on the same processor or different processors, and regardless of whether different computer processes run under the same operating system process / thread or different operating system processes / threads.

[0218] Computer programs may be permanently or transiently fixed in any form (e.g., source code, computer executable, or intermediate form) within a tangible storage medium, such as semiconductor memory elements (e.g., RAM, ROM, PROM, EEPROM, or flash programmable RAM), magnetic memory elements (e.g., diskettes or fixed disks), optical memory elements (e.g., CD-ROMs), PC cards (e.g., PCMCIA cards), or other memory elements.

[0219] Computer programs may be fixed in any form of signal that can be transmitted to a computer using any of the various communication technologies, including but not limited to analog technology, digital technology, optical technology, wireless technology (e.g., Bluetooth®), networking technology, and internetworking technology.

[0220] Computer programs may be distributed in any form as removable storage media (e.g., shrink-wrapped software) with printed or electronic documents, pre-loaded into computer systems (e.g., system ROM or fixed disk), or distributed from servers or electronic bulletin boards via communication systems (e.g., the Internet or the World Wide Web).

[0221] Hardware logic (including programmable logic used with programmable logic devices) that implements all or some of the functionalities described herein may be designed using conventional manual methods, or may be designed, captured, simulated, or electronically documented using various tools such as computer-aided design (CAD), hardware description languages ​​(e.g., VHDL or AHDL), or PLD programming languages ​​(e.g., PALASM, ABEL, or CUPL).

[0222] Any suitable computer-readable medium may be used. This computer-readable medium may, for example, be an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, device, or medium, but is not limited to these.

[0223] More specific examples of computer-readable media include, but are not limited to, electrical connections with one or more wires, or other tangible storage media such as portable computer diskettes, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), compact disc read-only memory (CD-ROM), or other optical or magnetic memory elements.

[0224] The programmable logic may be permanently or transiently fixed within a tangible storage medium, such as a semiconductor memory element (e.g., RAM, ROM, PROM, EEPROM, or flash programmable RAM), a magnetic memory element (e.g., a diskette or fixed disk), an optical memory element (e.g., a CD-ROM), or other memory element.

[0225] Programmable logic may be fixed in the form of signals that can be sent to a computer using any of the various communication technologies, including but not limited to analog technology, digital technology, optical technology, wireless technology (e.g., Bluetooth®), networking technology, and internetworking technology.

[0226] The programmable logic may be distributed as a removable storage medium (e.g., shrink-wrapped software) with printed or electronic documents and pre-loaded into a computer system (e.g., system ROM or fixed disk), or distributed from a server or electronic bulletin board via a communication system (e.g., the Internet or the World Wide Web). Naturally, some embodiments may be implemented as a combination of both software (e.g., computer program products) and hardware. Still other embodiments may be implemented entirely as hardware or entirely as software.

[0227] While specific exemplary embodiments have been described and illustrated in the accompanying drawings, it should be understood that these embodiments are illustrative and not limited to the specific structures and configurations illustrated and described, as various other changes, combinations, omissions, modifications, and substitutions are possible in addition to those described in the paragraphs above.

[0228] Those skilled in the art will recognize that various adaptations, modifications, and / or combinations of the embodiments described herein can be constructed. Therefore, it should be understood that within the scope of the appended claims, this disclosure may be practiced in ways other than those specifically described herein. For example, unless otherwise expressly mentioned, the steps of the processes described herein may be performed in a different order than described herein, and one or more steps may be combined, divided, or performed simultaneously. Furthermore, those skilled in the art will recognize that, in light of this disclosure, different embodiments or aspects described herein may be combined to form other embodiments.

Claims

1. An autonomous vehicle management system that dynamically manages the driving routines of multiple autonomous vehicles (V), An external control center (300; 300') having at least an exclusive area manager (304; 304'), One or more autonomous vehicles (V) are connected to the external control center (300; 300') in a communicative manner, The system comprises the external control center (300; 300') or a route planner (302; 302') included in one or more autonomous vehicles (V), The route planner (302; 302') is configured to calculate an initial route (IR) and assign it to one or more autonomous vehicles (V), and the exclusive area manager (304; 304') is configured to calculate and store an exclusive safety area (SA) for the initial route (IR), The one or more autonomous vehicles (V) are configured to move autonomously from a predetermined starting point to a destination point based on the assigned initial route (IR), The external control center (300;300') and / or the one or more autonomous vehicles (V) are further configured to detect changes in the environment around the initial path (IR) and to determine at least one of the potential or actual hazards to the one or more autonomous vehicles (V) moving along the assigned initial path (IR), If the changes in the environment around the assigned initial route are determined to be a potential hazard, the external control center (300;300') and / or one or more autonomous vehicles (V) are further configured to perform a route candidate generation process, wherein the route candidate generation process includes at least: A step of calculating, for a first autonomous vehicle among the one or more autonomous vehicles (V), at least one route candidate (CR) different from the assigned initial route (IR), and an exclusive safe area candidate (CSA) associated with the at least one route candidate, The steps include storing at least one set of route candidates (CRs) and exclusive safe area candidates (CSAs) associated with the route candidates (CRs), If the change in the environment around the initial path is determined to be an actual danger, the first autonomous vehicle is configured to perform a path switching process, and the path switching process is at least The steps include selecting one of the stored route candidates (CRs) and associated exclusive safety area candidates, The steps include setting the selected route candidate (CR) as the new initial route, and setting the exclusive safe area candidate (CSA) associated with the selected route candidate as the new exclusive safe area, An autonomous vehicle management system, comprising the step of continuously moving along the aforementioned new initial path.

2. The route candidate generation process performed by the external control center (300; 300') and / or the one or more autonomous vehicles (V) further, The autonomous vehicle management system according to claim 1, comprising adding the stored exclusive safety area candidates (CSAs) to the areas assigned to the exclusive safety area (SA) of the first autonomous vehicle.

3. The route candidate generation process performed by the external control center (300; 300') and / or the one or more autonomous vehicles (V) further, Sending a request for each of the calculated route candidates (CRs) to the exclusive area manager (304; 304') to store each of the calculated route candidates (CRs) and calculated exclusive safe area candidates (CSAs), The autonomous vehicle management system according to claim 1 or 2, further comprising storing at least one of the set of route candidates (CRs) and exclusive safe area candidates (CSAs) only when a storage permission message is received from the exclusive area manager (304; 304').

4. The exclusive area manager (304; 304') of the external control center (300; 300') is configured to store information about each exclusive safe area (SA) currently stored by the external control center (300; 300'), and each exclusive safe area candidate (CSA) currently stored by the external control center (300; 300') and / or one or more autonomous vehicles (V), The autonomous vehicle management system according to claim 3, wherein if the requested exclusive safety area candidate (CSA) does not overlap with at least any of the stored exclusive safety areas (SAs) and the stored exclusive safety area candidate (CSA), the exclusive area manager (304; 304') is configured to access the stored information and transmit the storage permission message to the first autonomous vehicle.

5. The external control center (300; 300') and / or the one or more autonomous vehicles (V) are configured to determine at least one of the potential hazards or the actual hazards based on a hazard determination process, and the hazard determination process is The steps include receiving object information relating to an object (O) identified as a potential or actual hazard, which includes at least the position of the object (O), or the position and velocity of the object (O); The autonomous vehicle management system according to claim 1 or 2, comprising the step of determining whether the object (O) is a potential hazard or an actual hazard to one of the one or more autonomous vehicles (V) by comparing the received object information with a predetermined threshold defined by each of the one or more autonomous vehicles (V).

6. The predetermined threshold is defined by at least the distance from the object (O) to one of the one or more autonomous vehicles (V), The object (O) is defined as a potential hazard if the distance from the object (O) to at least one of the one or more autonomous vehicles (V) is less than a predetermined potential hazard value and greater than a predetermined actual hazard value. The autonomous vehicle management system according to claim 5, wherein the object (O) is defined as an actual hazard if the distance from the object (O) to at least one of the one or more autonomous vehicles (V) is less than a predetermined actual hazard value.

7. The at least one route candidate (CR) calculated by the first autonomous vehicle is calculated to avoid the potential hazard caused by an object (O) to the first autonomous vehicle, The autonomous vehicle management system according to claim 1 or 2, wherein the external control center (300;300') and / or the one or more autonomous vehicles (V) are configured to calculate the at least one route candidate (CR) based on the relative position of the object (O) with respect to the first autonomous vehicle.

8. The at least one route candidate (CR) calculated by the external control center (300;300') and / or the one or more autonomous vehicles (V) is calculated to avoid the potential hazard caused by an object (O) for the first autonomous vehicle. The autonomous vehicle management system according to claim 1 or 2, wherein the first autonomous vehicle is configured to calculate the at least one route candidate (CR) based on a prediction of the movement of the object (O).

9. The step of calculating the at least one route candidate (CR) by the external control center (300;300') and / or the one or more autonomous vehicles (V) includes at least the calculation of a first route candidate (CR1) and a second route candidate (CR2), The first candidate route (CR1) is defined to maintain the maximum distance from the first autonomous vehicle to the potential hazard. The autonomous vehicle management system according to claim 1 or 2, wherein the second candidate route (CR2) is defined to maintain a predefined distance from the first autonomous vehicle to the potential hazard and ultimately align with the initial route.

10. The external control center (300; 300') and / or the one or more autonomous vehicles (V) are configured to manage the stored route candidates (CR), The autonomous vehicle management system according to claim 1 or 2, wherein the external control center (300; 300') and / or the one or more autonomous vehicles (V) are configured to erase and / or update stored route candidates (CRs) and associated exclusive safe area candidates (CSAs) based at least on detected changes in the environment or after a predetermined time has elapsed since storage.

11. The exclusive area manager (304; 304') of the external control center is configured to store information regarding each stored exclusive safe area (SA) and each stored exclusive safe area candidate (CSA), If the requested exclusive safe area candidate (CSA) overlaps only with the stored initial route (IR), the exclusive area manager (304; 304') is configured to send the storage permission message to the first autonomous vehicle. The autonomous vehicle management system according to claim 3, wherein the route planner (302; 302') is configured to erase the stored initial route (IR) that overlaps with the exclusive safety area candidate (CSA), and to reassign the new initial route to the one or more autonomous vehicles (V) to which the erased initial route was previously assigned.

12. If, after the external control center (300;300') and / or one or more autonomous vehicles (V) have sent a request to the exclusive area manager (304;304') to store each calculated route candidate (CR) and exclusive safety area candidate (CSA) for a calculated route candidate (CR), the exclusive area manager (304;304') has not sent a storage permission message to the first autonomous vehicle, the route planner (302;302') is configured to calculate an alternative route candidate, and the exclusive area manager (304;304') is configured to calculate an alternative safety area candidate associated with the alternative route candidate. The exclusive area manager (304; 304') is configured to send the alternative route candidates and the alternative safety area candidates to the first autonomous vehicle. The autonomous vehicle management system according to claim 3, wherein the first autonomous vehicle is configured to store the alternative route candidate and the alternative safety area candidate as route candidate (CR) and the exclusive safety area candidate (CSA) associated with the route candidate (CR).

13. The external control center (300; 300') and / or the one or more autonomous vehicles (V) are configured to store map data including two-dimensional information relating to the area in which the one or more autonomous vehicles (V) are moving. The two-dimensional information includes at least information relating to one or more vehicle priority areas (PAs) that define the travel area where it is preferable for one or more autonomous vehicles to travel, and one or more vehicle restriction areas (VPAs) that define the travel area where it is prohibited for one or more autonomous vehicles to travel. The route planner (302; 302') is configured to access the map data and calculate and assign an initial route (IR) by using only the locations defined by the vehicle priority area (PA) for route calculation. The autonomous vehicle management system according to claim 1 or 2, wherein the exclusive area manager (304; 304') of the external control center (300; 300') is configured to access the map data and calculate and store the exclusive safe area (SA) of the corresponding initial route (IR) by using only the locations defined by the vehicle priority area (PA) for exclusive safe area calculation.

14. The autonomous vehicle management system according to claim 13, wherein the exclusive area manager (304; 304') is configured to transmit the stored transmission message to the external control center (300; 300') and / or the one or more autonomous vehicles (V) by comparing the requested exclusive safe area candidate (CSA) with the map data and transmitting the stored transmission message if the requested exclusive safe area candidate (CSA) overlaps with an area defined by the vehicle priority area (PA) and / or the vehicle prohibition area (VPA).

15. A method for dynamically managing the driving routines of a plurality of autonomous vehicles (V) by an autonomous vehicle management system, comprising at least an external control center (300; 300'), one or more autonomous vehicles (V) communicably connected to the external control center (300; 300'), and a route planner (302; 302'), wherein the external control center (300; 300') further comprises at least an exclusive area manager (304; 304'), and the route planner (302; 302') is included in the external control center (300; 300') or the one or more vehicles (V), wherein at least, The steps include: calculating an initial route (IR) using the route planner (302; 302') and assigning it to one or more autonomous vehicles (V); and calculating and storing an exclusive safety area (SA) for the initial route (IR) using the exclusive area manager (302; 204'); The steps include: one or more autonomous vehicles (V) autonomously moving from a predetermined starting point to a destination point based on the assigned initial route (IR); The steps include: detecting changes in the environment around the initial path (IR) using the external control center (300;300') and / or the one or more autonomous vehicles (V), and determining at least one of a potential hazard or an actual hazard to the one or more autonomous vehicles (V) moving along the assigned initial path (IR); If the change in the environment around the assigned initial route (IR) is determined to be a potential hazard, the external control center (300;300') and / or one or more autonomous vehicles (V) calculate for a first autonomous vehicle among the one or more autonomous vehicles (V) at least one route candidate (CR) different from the assigned initial route (IR) and an exclusive safe area candidate (CSA) associated with the at least one route candidate (CR), and store the at least one route candidate (CR) and the exclusive safe area candidate (CSA), If the change in the environment around the initial path is determined to be an actual danger, the first autonomous vehicle selects one of the stored path candidates (CR), The first autonomous vehicle sets the selected route candidate (CR) as a new initial route, and sets the selected exclusive safe area candidate (CSA) associated with the selected route candidate (CR) as a new exclusive safe area. A method comprising the step of having the first autonomous vehicle continue to move along the new initial path.

Images