Security system network for monitoring area and method for monitoring such area

Abstract

A security system network (1) for monitoring an area (5), such as a warehouse or factory, in which objects (8), such as autonomously driven vehicles (4) and persons, move together. The security system network (1) comprises a central processing device (3) which is designed to receive sensor data (6) from a plurality of monitoring units (2), in which sensor data (6) objects (8) detected by the monitoring units (2) in a monitored area (5) are contained. The central processing device (3) is designed to merge the received sensor data (6). The central processing device (3) is designed to create an object list (7) from the merged sensor data, the object list (7) containing the detected objects (8) and the corresponding object information, and to transmit the object list (7) to the autonomously driven vehicle (4).

Description

Technical Field

[0001] The present invention relates to a security system network for monitoring an area and a method for monitoring the area. Background Art

[0002] Autonomous vehicles, especially forklifts that can be used in warehouses or logistics centers, can improve the efficiency and safety of processes in such warehouses. These autonomous vehicles are programmed to independently transport, stack, and sort goods and materials without the need for a human driver.

[0003] Autonomous vehicles can be used in various ways. In some areas, only autonomous vehicles are used, while in others, humans may also enter during normal operations, resulting in mixed operations. In these highly dynamic environments, humans and machines move within confined spaces, performing a wide range of tasks, from receiving goods to storage and shipping. For autonomous vehicles to effectively and safely integrate into these environments, advanced sensors and algorithms are required to recognize and interpret the environment. This helps avoid collisions and ensure smooth operation.

[0004] To avoid accidents, industrial environments use packaged safety components for low-complexity implementation. In autonomous vehicles, the typical safety chain, consisting of safety sensors, safety controllers, and safety actuators, primarily utilizes certified safety components with fixed functionality, defined safety levels, clearly described rule-based applications, and fixed interfaces. In these systems, sensor usage and hazard avoidance logic are determined during commissioning and remain unchanged during operation. However, minor dynamic modifications to safety functions are possible, such as in the form of field switching or muting. But generally speaking, current safety technology is static. This means that safety functions (such as field monitoring) execute as planned as long as the acquired sensor data is consistent. If the system detects flaws in the sensor data and inconsistencies with the expected process, a safe state is declared, requiring external intervention. In this case, the autonomous vehicle typically comes to a complete stop. Conventional safety technology is very rigid in this sense, aiming to avoid complexity through strict packaging and the use of certified safety components with fixed functionality.

[0005] This architecture has proven its worth for decades thanks to its simplicity and modular design. As automated processes become increasingly complex, simple safety concepts are often no longer sufficient. Traditional approaches are limited by the strong local focus of safety assurance (directed only at the hazard point itself), the low information content of sensor data, and a lack of flexibility to changing process conditions and error scenarios. In particular, the lack of resilience to error conditions and inconsistent or poor sensor data severely limits the productivity of such autonomous vehicles. In such autonomous vehicles, the use of safety technology is limited by the component's risk class. Therefore, it is not possible to dynamically add additional, independent safety information to ensure safety for risks in higher risk categories. Summary of the Invention

[0006] The object of the present invention is therefore to create a way to ensure the safety of complex automation processes in a wide range of operating environments and for a variety of hazardous scenarios. The present invention makes it possible to react dynamically and appropriately to changing situations. In particular, it makes it possible to maintain the safe operation of the automation process even when errors, inconsistencies, or poor quality affect the sensor data used.

[0007] This object is achieved by a safety system network for monitoring an area according to claim 1 and a corresponding method for monitoring the area according to claim 15. Advantageous further developments of the safety system network are specified in claims 2 to 14.

[0008] A safety system network is used to monitor areas, such as warehouses or factories, where objects, such as autonomous vehicles and people, move around together. The safety system network includes a central processing device configured to receive sensor data from multiple monitoring units, including objects detected by the monitoring units in the monitored area. The central processing device is configured to merge (i.e., integrate) the received sensor data. The central processing device is further configured to create object lists from the merged sensor data, wherein the object lists include the detected objects and corresponding object information. The central processing device is further configured to transmit these object lists to the autonomous vehicles.

[0009] It is particularly advantageous to combine (also called fusion) different sensor data in order to detect and appropriately characterize the various objects in the area to be monitored. It is also particularly advantageous to transmit this information to the autonomous vehicle in the form of an object list. In this case, the autonomous vehicle has additional information that it can use to assess various situations from a safety perspective. Furthermore, the use of a large number of monitoring units significantly increases the safety-related information content of the sensor data. This means that the autonomous vehicle can obtain information not only from the immediate surroundings of the hazardous point, but also from larger areas (entire factories, warehouses, etc.), and use this information to simultaneously achieve hazard avoidance and productivity improvements for multiple hazardous points.

[0010] Preferably, an object list is created for each object, which may be a person, a stationary object (such as a pallet) or an autonomous vehicle. Each object list includes at least one object information.

[0011] In particular, the safety system network according to the present invention decouples the rigid connection between data acquisition and processing (e.g., filtering) and the use of that data. This makes the safety chain more flexible, allowing, on the one hand, the dynamic integration of sensor data based on availability and suitability, and, on the other hand, the dynamic use of object information in different functions (particularly in autonomous vehicles). By using the object information received via object lists, the safety system of an autonomous vehicle can respond more flexibly and adaptably to hazardous situations. In particular, if errors, poor quality, or inconsistent data are detected in various sensor sources, this redundant information can be used to switch to a fallback operation that preserves productivity, if necessary.

[0012] The safety system network according to the present invention thus allows information from different sources (e.g. sensor data) to be combined (i.e. fused). Consequently, in particular in autonomous vehicles, higher-quality, more dynamic functions can be implemented than before. Furthermore, proactive risk avoidance can be achieved. It is also possible to create a central information node in the form of a central processing device, which is designed, for example, in a cloud system. Furthermore, since more precise control of autonomous vehicles is possible due to the additional information (object lists containing object information), slow driving of these autonomous vehicles is reduced, thereby optimizing the productivity of the autonomous operating processes of the autonomous vehicles. In particular, downtime of such autonomous vehicles can also be reduced. The safety system network according to the present invention also allows the connection of safety functions and automation functions, as well as the integration of the corresponding data in IT / OT systems.

[0013] In another embodiment, the object information of an object includes the position, size, movement direction and / or movement speed, object ID and / or object category of the object. The object category indicates whether the object is, for example, a person, an autonomous vehicle or a stationary object. The object information can be extracted from sensor data, which is generated by multiple monitoring units. The object information can be generated over a longer period of time or once. Preferably, the object information is continuously updated by the multiple monitoring units. In particular, the central processing device is designed to extract object information from the multiple sensor data. The central monitoring unit is designed to collect sensor data and, in particular, to evaluate the information content of the sensor data, i.e., the object information contained in the sensor data.

[0014] In another embodiment, the central processing unit is designed to convert the sensor data from multiple monitoring units into a common spatial and temporal coordinate system. This is a method for merging sensor data. The common spatial and temporal coordinate system can cover the entire monitored area or a portion of the area.

[0015] In another embodiment, the central processing unit is configured to graphically display the common spatial and temporal coordinate system on an output unit (e.g., a screen and / or a webpage). Different objects can be highlighted in different ways, for example, by color, shading, symbols, and / or size. In particular, objects with a confidence level below a threshold can be highlighted, thereby enabling a user viewing the output unit to identify a potential dangerous situation.

[0016] In another embodiment, the central processing device is designed to perform a plausibility check on the received sensor data by checking whether an object is contained in the sensor data of at least two monitoring units (the monitoring areas of the at least two monitoring units at least partially overlap (spatially)) and / or whether an object that moved relative to different but adjacent time periods is contained in the sensor data of at least two monitoring units (the monitoring areas of the at least two monitoring units are adjacent to each other). If this is the case, a higher trust level can be assigned to the object than if the object is contained in the sensor data of only one monitoring unit. In the latter case, a maintenance message can optionally be issued, i.e., at least one of the monitoring units should be checked for its normal functionality. In this case, the central processing device is designed not only to collect and merge a large amount of sensor data from the monitoring units, but also to perform a plausibility check on the sensor data.

[0017] In another embodiment, the safety system network includes at least one evaluation device designed to determine a trust level for an object based on sensor data and / or object information. Preferably, the evaluation device is located in a central processing unit. In this case, the central processing unit is not only responsible for collecting and merging large amounts of sensor data from monitoring units, but also for evaluating the sensor data regarding its information content and determining a trust level for each object. This trust level is transmitted to at least one autonomous vehicle along with or within a list of objects. The autonomous vehicle that receives the trust level for an object is designed to provide different safety functions based on the trust level. If the trust level of an object in the object's surroundings (within a predetermined distance range) is low, the autonomous vehicle may stop; if the trust level is high, the autonomous vehicle may continue driving without reducing speed, even if the object is in close proximity (within a predetermined distance range) to the autonomous vehicle. Alternatively, the evaluation device may also be located in one or more autonomous vehicles. In this case, object information is transmitted from the central processing unit (fusion core) to the autonomous vehicle without any safety evaluation. In principle, it is particularly preferred that the safety system network carries out a safety-related evaluation and classification of the detected objects.

[0018] In another embodiment, the evaluation device is designed to determine a safety degree, a safety level, and / or a performance level based on a trust level or based on sensor data and / or object information. For example, there are performance levels a to e. Performance levels represent the probability of a dangerous failure per hour ((PFHd) 1 / h). They are defined in the standard document ISO 13849-1:2006. Reference is also made to the document IEC 61508. Additionally or alternatively, the evaluation device is further designed to determine a safety integrity level (SIL) from SIL1 to SIL4 based on a trust level or based on sensor data and / or object information.

[0019] In another embodiment, the evaluation device is configured to assign a higher confidence level to an object if the object is contained in sensor data generated at the same time by at least two monitoring units, and if the monitoring fields of the at least two monitoring units at least partially overlap. In this case, object information such as position, size, direction of movement, and speed of movement can be verified based on the sensor data of the at least two monitoring units.

[0020] In a further embodiment, the evaluation device is designed to determine a trust level for the object as a function of the quality of the sensor data and / or the object information.

[0021] In another embodiment, the quality of the sensor data depends on the physical characteristics of each monitoring unit. These physical characteristics include, in particular, the age, type, error rate, failure rate, scattering rate, measurement method, installation location, and / or trust information of each monitoring unit. A newly installed monitoring unit can be assumed to have a low failure rate. If the installation location has a uniform temperature and the monitoring unit is not affected by precipitation, the quality is higher than if the monitoring unit is exposed to severe temperature fluctuations and precipitation. Monitoring units with a good signal-to-noise ratio have better quality than those with a poor signal-to-noise ratio. If the scattering rate of a monitoring unit's sensor data is high, the quality is poor. Additionally or alternatively, the quality of the object information depends on the location, direction of movement, and / or object type. Of particular importance is whether the object is a person or an autonomous vehicle. For example, in the case of an autonomous vehicle, it can be assumed that the direction of movement corresponds to a specific motion profile, but this is not always true in the case of a person. The speed of an autonomous vehicle is also more constant than that of a person.

[0022] In another embodiment, the better the quality of the sensor data, the higher the trust level.

[0023] In another embodiment, the evaluation device comprises an AI module (KI-Modul). The AI ​​module is designed to determine a trust level for an object based on sensor data and / or object information. In principle, such an AI module can be trained using already recorded sensor data. Sensor data from a large number of monitoring units can be input to the AI ​​module. In principle, it is conceivable that, in addition to the sensor data, the AI ​​module is also given information about what type of object the object is (person, autonomous vehicle, stationary object) and how to evaluate the trust level of each object. Based on the movement of the classified objects, the AI ​​module can recognize the difference between a person and an autonomous vehicle. Based on these training data, the trained AI module can independently determine the trust level of other objects based on other sensor data recorded by other monitoring data. It is also conceivable that scenarios leading to abnormal situations and / or accidents are input to the AI ​​module for training.

[0024] In another embodiment, a safety system network includes at least one autonomous vehicle, wherein the at least one autonomous vehicle is configured to receive a list of objects from a central processing device.

[0025] In another embodiment, the central processing device is configured to transmit to the autonomous vehicle only an object list containing objects within a predetermined distance from the autonomous vehicle and corresponding object information. In this case, the object list including all objects within the monitored area is not transmitted to the autonomous vehicle.

[0026] In another embodiment, the at least one autonomous vehicle is configured to enter the intersection without braking if an object in the object list whose distance from the intersection is less than a distance threshold has a trust level greater than a first threshold. Conversely, the at least one autonomous vehicle is configured to enter the intersection at a reduced speed or stop if an object in the object list whose distance from the intersection is less than the distance threshold has a trust level less than a second threshold. The first threshold and the second threshold may be the same or different.

[0027] This preferred embodiment contributes to safety at intersections, particularly in industrial environments. The danger lies in the fact that at such intersections, the paths of people and autonomous vehicles can intersect. Due to the restricted field of view at such intersections, it is not always possible to proactively detect dangerous situations using the safety sensors installed on the autonomous vehicles. Therefore, autonomous vehicles have previously had to travel through intersections at low speeds. This embodiment thus avoids the associated reduction in productivity, as sensor data is already provided by other sources, with which it is possible to reliably determine whether a person is located in the intersection area or is approaching it.

[0028] The following describes how to implement this application scenario (intersection entry). Using optical and / or wireless positioning sensors in monitoring units, the positions of all relevant objects (e.g., people and autonomous vehicles in the monitored area) are recorded and compiled into a common real-time map. A real-time map can be understood as the registration of all recorded objects in a spatial and temporal coordinate system. For example, a 3D sensor with an object tracking algorithm can be used here. Additionally or alternatively, a UWB positioning system with transponders installed on each person and vehicle can also be used. The object information obtained in the form of position data is merged into a central processing unit (fusion module) and checked for accuracy, consistency, trust information, and / or prior knowledge, improving the safety assessment (i.e., trust level) for each object. This list of "safe" objects forms the data basis for further use by all connected safety functions in each autonomous vehicle. For example, in intersection monitoring, different situations are distinguished and treated differently from a safety perspective. In the case of good and consistent measurement values, i.e. if, for example, the position data of the optical system and the UWB system match and agree with the historical trajectory within tolerances and a good confidence value is provided, the safety function can use the following logic for hazard avoidance:

[0029] - Decentralized safety systems on autonomous vehicles are being bypassed. In this case, it is called "silencing."

[0030] - If no one is present in the intersection area, the autonomous vehicle can pass through the intersection without reducing speed.

[0031] If a person approaches, a message can be sent to the person to temporarily avoid entering the intersection (via a message sent via the person tag). This message can be sent to the person, for example, via an optical and / or acoustic system. If the person does not approach further, the autonomous vehicle proceeds through the intersection at full speed. If a person approaches, the autonomous vehicle reduces its speed and, if necessary, stops until the pedestrian has left the intersection.

[0032] If the safety level (i.e., trust level) of an object present in the relevant intersection environment is assessed by the central processing device to be low, a more cautious safety assurance logic is used:

[0033] - The autonomous vehicle's decentralized safety system remains active and, when approaching an intersection, it will choose to reduce speed and stop at the intersection. In this case, productivity is limited, but autonomous operation is still possible.

[0034] If a decentralized safety system in an autonomous vehicle fails, in a well-known classic case, this would result in an emergency stop of the autonomous vehicle. Autonomous operation can only be resumed after the fault is corrected through external measures. In the advanced safety system network described here, the autonomous vehicle's operation can be maintained using redundant object information from a safe object list. As long as the generated object information is of sufficient quality and safety capability (trust level), the autonomous vehicle continues to operate. At the same time, the fault can be reported, and troubleshooting can be scheduled at a convenient time.

[0035] This safety assurance logic can be implemented as a cascade of safety functions for hazardous points at intersections. Similar to logic blocks in a programmable safety controller, in this case, the safety function blocks are linked by event differentiation and depending on the quality of the object information, for optimal and robust operation.

[0036] In another embodiment, at least one autonomous vehicle is configured such that, in the event of a vehicle entering an intersection without braking, at least one or all of its safety systems (e.g., laser scanners, etc.) are bypassed or deactivated. In this case, the safety system causing braking or an emergency stop is not triggered.

[0037] In another embodiment, at least one autonomous vehicle is designed to travel within a heavy-duty truck (LKW) and / or rail vehicle to load or unload cargo. The autonomous vehicle only enters the LKW and / or rail vehicle if it is within a distance range of the LKW and / or rail vehicle that is less than a distance threshold, an object in the object list is not a person, and the trust level of the object's corresponding object information is greater than a threshold. The autonomous vehicle is designed to bypass or deactivate one or all safety systems when entering the LKW and / or rail vehicle. This ensures that the autonomous vehicle can load or unload cargo in the tightest of spaces without braking or emergency stops. This further embodiment may also be referred to as advanced vehicle safety.

[0038] The following provides further details on how this vehicle safety assurance is implemented. Assume that autonomous vehicles are parked in different operational areas of a factory or warehouse. For example, they receive goods directly from a transfer point in a truck, travel from there via the facility's central lane to a separate warehouse, and deliver the received goods to a robot, particularly a sorting robot. Depending on the area and situation, the safety assurance of the autonomous vehicle is implemented in completely different ways. The position data from the aforementioned real-time map is used so that optimal productivity is achieved, preferably taking into account the autonomous vehicle's current position and surroundings at each point in time.

[0039] If the autonomous vehicle is in the transfer area of ​​a truck or rail vehicle, the autonomous vehicle's advanced safety functions use the derived Safe Point of Interest function and the configured area to detect: in this case, due to the narrow conditions and obstructions, safety must be guaranteed without the use of on-board safety systems (such as safety scanners). Only the available positioning information from the real-time map is used here. Specifically, the autonomous vehicle monitors the immediate surroundings and, in particular, the areas not visible from the truck or rail vehicle via a central processing unit or by receiving an object list from the central processing unit for the presence of people. If the position information is missing or implausible, or if a malfunction occurs, the autonomous vehicle stops. If valid position information is available and no approach is detected, the autonomous vehicle can approach the truck or rail vehicle and take over its cargo.

[0040] In the case of goods delivered to warehouses, autonomous vehicles are often deployed in areas frequently frequented by people and other (autonomous) vehicles. In such situations, leveraging the autonomous vehicle's onboard safety systems (e.g., safety sensors) is essential. This provides the primary safety function for the area and, if necessary, can be supplemented by location data from real-time maps. This optimizes reliability, avoidance functions, intersection monitoring, and more.

[0041] Upon reaching the compartmented warehouse, the security function can switch back to location-based environmental monitoring based on the received object list. In this example, the warehouse is assumed to be fenced or walled off, with personnel only allowed access in the event of maintenance or breakdowns. When the autonomous vehicle enters through the gate, location data is used to monitor that no one has passed through the autonomous area of ​​the warehouse. The autonomous vehicle's safety systems are then bypassed (silenced), and the autonomous vehicle moves through the warehouse's storage area without engaging any safety mechanisms that could reduce productivity.

[0042] In all such cases, the advanced vehicle safety function obtains location information and objects in the autonomous vehicle's surroundings from an object list received from the central processing unit. Based on the location and its plausibility, particularly the confidence level, the advanced vehicle safety function selects the appropriate safety function for the situation (through which the autonomous vehicle's safety system operates), and executes the selected function based on the object information and confidence level.

[0043] In another embodiment, at least one autonomous vehicle is designed to continue driving even if at least one safety system for environmental monitoring fails if objects on the object list will not cause a collision and their confidence levels are greater than a threshold.

[0044] The method according to the invention is used for monitoring areas, such as warehouses or factories, in which objects such as autonomous vehicles and people move together. In a first method step, sensor data is received from a plurality of monitoring units, which contains objects detected by the monitoring units in the monitoring area. In a second method step, the received sensor data are merged. In a third method step, an object list is created from the merged sensor data, wherein the object list contains the detected objects together with the corresponding object information. In a fourth method step, these object lists are transmitted to the autonomous vehicles. Obviously, the transmission of the object lists to the autonomous vehicles also only includes the transmission of parts of the object lists.

[0045] In another embodiment, the single autonomous vehicle is a forklift, or the plurality of autonomous vehicles are forklifts.

[0046] In another embodiment, the central processing device may also be described as a sensor fusion module and functionally serves as the central interface for the safety system network.

[0047] In another embodiment, the central processing device is designed to collect (receive) sensor data from multiple monitoring units, perform rationality checks and filtering on the sensor data from multiple monitoring units, evaluate the information content of these sensor data (create object information and trust levels), and transmit these sensor data to the autonomous vehicle to perform various safety functions.

[0048] In another embodiment, the central processing unit is configured to provide the user with information regarding the presence of an object in their vicinity whose trust level falls below a threshold. This information can be communicated to the user, for example, via a radio message and / or visually, for example, by means of a corresponding signaling device near the user. Acoustic or tactile notifications (e.g., via vibration) are also conceivable.

[0049] In another embodiment, the central processing device is designed to create objects using object information in sensor data from different monitoring units. This allows for a more accurate description of the corresponding object and a higher level of trust.

[0050] In another embodiment, the at least one autonomous vehicle is configured to obtain an updated object list from the central processing device at regular intervals, in particular by downloading the updated object list from the central processing device. It is also possible that the central processing device sends the updated object list to the at least one autonomous vehicle at regular intervals.

[0051] In another embodiment, the transmission of the object list from the central processing device to the at least one autonomously driven vehicle occurs via a wireless communication standard (eg, WLAN).

[0052] In another embodiment, the monitoring unit is arranged (in particular installed) exclusively or predominantly in the area to be monitored.

[0053] In a further embodiment, at least two monitoring units are designed to generate partially overlapping monitoring fields.

[0054] In another embodiment, the driving route passes through at least a portion of the area to be monitored, wherein the autonomously driven vehicle moves exclusively along this driving route, at least in normal operating mode.

[0055] In a further embodiment, the monitoring units of the safety network are arranged in such a way that at least 80% or at least 90% of the driving route is always covered by the monitoring fields of at least two monitoring units.

[0056] In another embodiment, the monitoring unit is an optical positioning sensor and / or a wireless-based positioning sensor.

[0057] In another embodiment, the object list transmitted to the autonomously driven vehicle also includes object information specifically for the autonomously driven vehicle, such as position, movement direction and / or movement motion.

[0058] In another embodiment, the autonomously driven vehicle is designed to perform a check of its own measured values ​​(eg speed, position and / or movement direction) based on at least one piece of object information received by the autonomously driven vehicle about itself.

[0059] In another embodiment, the autonomous vehicle is designed to send information to a higher-level control device, for example, to a central processing unit, in which the higher-level control device is informed that at least one of its own measured values ​​does not match the received object information describing the autonomous vehicle itself. In this case, for example, maintenance is planned and the autonomous vehicle is removed from production operation.

[0060] In another embodiment, the monitoring unit is a camera or a camera sensor, a radar sensor, a 3D scanner, a wireless-based tracking system, a contact ring in the ground and / or around the locking device. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] The invention is described below purely by way of example with reference to the accompanying drawings.

[0062] Figure 1 An embodiment of a safety system network according to the present invention is shown, which includes a plurality of monitoring units, a central processing device, and an autonomously driven vehicle.

[0063] Figure 2A visualization of a safety system network is shown, which illustrates how various data are transmitted within the safety system network.

[0064] Figure 3 A real-time map of the area to be monitored is shown, in which the corresponding detected objects are marked by the central processing device, which describes a more efficient crossing of the intersection.

[0065] Figure 4 A real-time map of the area to be monitored is shown, in which the corresponding detected objects are marked by a central processing device, wherein a more efficient unloading or loading of a truck is described.

[0066] Figure 5 A method for monitoring an area, such as a warehouse or a factory, is shown.

[0067] Reference Signs List

[0068] Security System Network 1

[0069] Monitoring unit 2

[0070] Camera 2a

[0071] Radar sensor 2b

[0072] Locking device 2c

[0073] Door System 2D

[0074] Robot 2e

[0075] Wireless based tracking system 2f

[0076] Central processing unit 3

[0077] Autonomous vehicles 4

[0078] Area 5

[0079] Normal area 5a

[0080] Separation area 5b

[0081] Sensor Data 6

[0082] Object List 7

[0083] Subject 8

[0084] Person 8a

[0085] Tray 8b

[0086] Coordinate System 9

[0087] Service Terminal 10

[0088] Application 11

[0089] Evaluation device 12

[0090] AI Module 13

[0091] Intersection area 14

[0092] Heavy-duty vehicles 15

[0093] Method steps S1, S2, S3, S4 DETAILED DESCRIPTION

[0094] Figure 1 An embodiment of a safety system network 1 according to the present invention is shown, comprising a plurality of monitoring units 2, a central processing device 3, and an autonomous vehicle 4. The safety system network 1 is used to monitor an area 5, such as a warehouse or factory. The plurality of monitoring units 2 are each configured to monitor a specific portion of the area 5 and generate corresponding sensor data 6. The monitoring units 2 are configured to transmit the sensor data 6 to the central processing device 3. The central processing device 3 is configured to merge the received sensor data 6. The central processing device 3 creates an object list 7 based on the merged sensor data, wherein the object list contains detected objects 8. These objects 8 may include people 8a, autonomous vehicles 4, and stationary objects such as pallets 8b. Each object has corresponding object information, which is also included in the object list 7. This object information may include the object's location, size, direction and / or speed of movement, object ID, and / or object category. The corresponding object list 7 is then transmitted to the autonomous vehicle 4.

[0095] The monitoring unit 2 may be various devices that are designed to generate sensor data 6 containing the object 8. Figure 1 In the embodiment, the monitoring units 2 are cameras 2a, radar sensors 2b, locking devices 2c, door systems 2d, robots 2e and wireless-based tracking systems 2f, which are attached, for example, to persons and / or vehicles.

[0096] The sensor data 6 can be raw data generated by each monitoring unit 2. It is also possible that the monitoring unit 2 has already processed the sensor data 6 so that the central processing device 3 can more easily process it. The sensor data 6 can take various forms. If the monitoring unit 2 is a camera 2a, the camera image can be transmitted directly to the central processing device 3 in the form of sensor data 6. It is also conceivable that the camera 2a itself detects an object 8 in the image and only transmits the object 8's position in the image or in the monitoring area 5 to the central processing device 3 in the form of coordinates. The above description also applies if the monitoring unit 2 is a radar sensor 2b. If the monitoring unit 2 is a locking device 2c, the activation of the locking device 2c can be detected and transmitted to the central processing device 3 in the form of sensor data. The central processing device 3 then detects that a person 8a or an autonomously driven vehicle 4 has left or entered the door area protected by the locking device 2c. The above description also applies if the monitoring unit 2 is an (automatic) door system 2d. If the corresponding door leaf opens and closes, this can be interpreted as an indication that a person 8a or an autonomous vehicle 4 has entered or left the door area. If the monitoring unit 2 is a machine, such as a robot 2e, the corresponding sensor data 6 transmitted from the machine to the central processing device 3 can reflect the machine's status. If the machine has completed a cycle or work sequence, this information can be interpreted as an indication that a person 8a or an autonomous vehicle 4 will soon arrive to receive the finished product or provide material to be processed. The monitoring unit 2 can also be a wireless tracking system 2f, for example, attached to a vehicle or person 8a to reliably record their position in the monitored area 5. The monitoring unit 2 can also be an autonomous vehicle 4 equipped with corresponding sensors, such as LIDAR sensors, to scan the environment.

[0097] Each monitoring unit 2 can be connected to the central processing device 3 via a wired connection or a wireless connection.

[0098] The central processing device 3 may be, for example, a central computer system arranged centrally or decentralized (eg in the cloud).

[0099] The central processing device 3 is designed to transform the sensor data 6 of the plurality of monitoring units 2 into a common spatial and temporal coordinate system 9. This common coordinate system 9 in the form of a real-time map is Figure 3 and Figure 4 In display.

[0100] The common coordinate system 9 can also be displayed on a service terminal 10, which is connected to the central processing device 3. Through the service terminal 10, control commands can be transmitted to the central processing device 3 and, more preferably, to the monitoring unit 2 connected to the central processing device 3 and / or the autonomously driven vehicle 4.

[0101] Figure 2 A visualization of a safety system network 1 is shown, illustrating how different data are transmitted within the safety system network 1. In this case, the visualization takes place on a service terminal 10. An application 11 is started on the service terminal 10 to control the central processing device 3. Sensor data 6 are transmitted from the individual monitoring units 2 to the central processing device 3. In the central processing device 3, the sensor data 6 are merged, wherein the central processing device 3 is designed to record each object 8 in the sensor data 6 or the merged sensor data and to create a corresponding object list 7. The object list 7 includes the individual objects 8 and their object information. The object list 7 is then transmitted to the autonomous vehicles 4. Not all autonomous vehicles 4 necessarily receive the same object list 7. An object list 7 contains at least one object 8 and at least one piece of object information for this object 8.

[0102] Preferably, the central processing device 3 is also configured to perform a plausibility check on the received sensor data 6. If the monitoring areas of at least two monitoring units 2 overlap, the object 8 must be detected in both sets of sensor data 6. Otherwise, the sensor data 6 is not plausible. The same applies if the monitoring areas of two monitoring units 2 are adjacent to each other. If an object 8 moves through two monitoring areas successively, it must be visible in both sets of sensor data 6 from the monitoring units 2, even if there is a temporal offset.

[0103] exist Figure 2 , it is also shown that the safety system network 1 includes an evaluation device 12. In this case, the evaluation device 12 is part of the central processing device 3. The evaluation device 12 is designed to determine a trust level for the object 8 based on the sensor data 6 and / or the object information. The evaluation device 12 is designed to determine a high trust level for the object 8 if the object 8 is contained in the sensor data 6 from at least two monitoring units 2 generated simultaneously and the monitoring fields of the at least two monitoring units 2 at least partially overlap, or if the object 8 appears in the sensor data 6 of different monitoring units 2 with a time offset and the monitoring fields of the different monitoring units 2 are arranged adjacent to each other.

[0104] exist Figure 26 also shows that the evaluation device 12 includes an AI module 13. The AI ​​module 13 is designed to determine a trust level for the object 8 based on the sensor data 6 and / or the object information. The central processing unit 3 is designed to add the trust level to the object list of the corresponding object 8 or directly to the object information for the object 8. The autonomous vehicle 4 is designed to respond adaptively to the trust level in different situations based on the object 8 in the object list and the trust level of the object 8.

[0105] Figure 3 A real-time map of the area 5 to be monitored is shown, in which the corresponding detected objects 8 are marked by the central processing device 3. Preferably, a specific trust level is determined for each object 8. The area 5 comprises a normal area 5a and a separation area 5b. The area 5a and the area 5b are separated from each other by a dotted line. In the normal area 5a, the autonomous vehicle 4 and the person 8a are in normal operation at the same time. In the separation area 5b, only the autonomous vehicle 4 is in normal operation. There are multiple monitoring units 2 in both the normal area 5a and the separation area 5b. The autonomous vehicle 4 is designed to deactivate (power off, mute, etc.) its safety system (e.g., laser scanner) in the separation area 5b and rely only on the object information of the object list 7, which is transmitted to the autonomous vehicle 4 via the central processing device 3. As a result, in particular, the trays 8b can be arranged closely to each other, since the safety system of the autonomous vehicle 4 does not trigger an emergency activation In contrast, preferably, the safety system is activated in the normal area 5a. In the normal area 5a, an intersection area 14 is also shown. The intersection area 14 and the surroundings of the intersection area 14 are monitored by the corresponding monitoring unit 2. Figure 2 In the example, autonomous vehicle 4, which has just entered intersection area 14 from below, has received object list 7 containing relevant objects 8 within intersection area 14 (objects within a certain distance from intersection area 14). The confidence level of objects 8 within intersection area 14 is greater than a threshold, allowing autonomous vehicle 4, which has just entered from below, to reliably determine that no other objects 8 are present in intersection area 14. Therefore, autonomous vehicle 4, which has just entered from below, is designed to enter intersection area 14 without reducing speed and, optionally, even deactivate its safety systems. If the confidence level is below a (e.g., different) threshold, autonomous vehicle 4, which has just entered from below, enters intersection area 14 at a reduced speed, or even stops before entering.

[0106] Figure 4Another real-time map of the monitored area 5 is shown, in which the corresponding detected objects 8 are marked by the central processing device 3. In this embodiment, it is explained how an autonomously driven vehicle 4 can more efficiently unload or load a truck 15. The autonomously driven vehicle 4 again receives an object list 7 from the central processing device 3, which contains a plurality of objects 8 and corresponding object information. A trust level is again generated for each object 8. The monitored area 5 includes a normal area 5a and a separated area 5b. Furthermore, the monitored area 5 is monitored by multiple monitoring units 2. At least one of these monitoring units 2 is arranged so that it can observe the truck 15 to be unloaded. The corresponding sensor data 6 of the at least one monitoring unit 2 may include objects 8 located within the truck 15 to be unloaded or loaded and / or objects 8 located directly near the truck 15 to be unloaded or loaded. These objects 8 can again be added to the object list 7 by the central processing device 3. Similarly, corresponding trust levels are generated for these objects 8 (e.g., based on the quality with which the corresponding monitoring unit is operated) and added to the object list 7. Autonomous vehicle 4 receives object list 7 and is configured to drive into truck 15 if information from object list 7 indicates that no person 8 a is within a certain distance threshold from truck 15 and the confidence level of the detected object 8 is greater than the threshold. In this case, autonomous vehicle 4 is configured to drive into truck 15 and bypass or deactivate one or all of autonomous vehicle 4's safety systems, such as the laser scanner. This prevents emergency activation of autonomous vehicle 4's safety systems due to the limited space within truck 15.

[0107] The central processing device 3 is also designed to send the object list 7 to a mobile terminal device of a person 8a, such as a smartphone. The person 8a then has real-time data on the current position of each object 8 in the area 5 to be monitored.

[0108] Figure 5 A method for monitoring an area 5, such as a warehouse or factory, is described. In a first method step S1, sensor data 6 are received from a plurality of monitoring units 2, including objects 8 detected by the monitoring units 2 in the monitoring area 5. In a second method step S2, the received sensor data 6 are merged. In a third method step S3, an object list 7 is created based on the merged sensor data, wherein the object list 7 contains the detected objects 8 together with the corresponding object information. In a fourth method step S4, these object lists are transmitted to an autonomously driven vehicle 4.

[0109] The invention is not limited to the described exemplary embodiments. Within the scope of the invention, all described and / or illustrated features can be combined with one another in any desired manner.

Claims

1. A security system network (1) for monitoring an area (5) in which objects (8) including autonomously driven vehicles (4) and people (8a) move together, in, The security system network (1) comprises a central processing device (3) which is designed to receive sensor data (6) from a plurality of monitoring units (2), wherein the sensor data (6) includes the objects (8) detected by the monitoring units (2) in the monitored area (5), wherein the central processing device (3) is designed to combine the received sensor data (6), The central processing device (3) is designed to create an object list (7) based on the merged sensor data, wherein the object list (7) contains the detected objects (8) and corresponding object information, and transmit the object list (7) to the autonomously driven vehicle (4).

2. The security system network (1) according to claim 1, characterized in that The object information of the object (8) includes a position, a size, a moving direction and / or a moving speed, an object ID and / or an object category for the object (8).

3. The security system network (1) according to claim 1 or 2, characterized in that The central processing device (3) is designed to transform the sensor data (6) of a plurality of the monitoring units (2) into a common spatial and temporal coordinate system.

4. The security system network (1) according to claim 1, characterized in that The central processing device (3) is designed to perform a plausibility check on the received sensor data (6) by performing the following checks: a) checking whether the object (8) is contained in the sensor data (6) of at least two monitoring units (2), wherein the monitoring areas of at least two monitoring units (2) at least partially overlap; and / or b) checking whether the sensor data (6) of at least two monitoring units (2) each contain the object (8) that moved relative to different but adjacent time periods, wherein the monitoring areas of at least two monitoring units (2) are adjacent to each other.

5. The security system network (1) according to claim 1, characterized in that The safety system network (1) comprises an evaluation device (12) which is designed to determine a trust level for the object (8) based on the sensor data (6) and / or the object information.

6. The security system network (1) according to claim 5, characterized in that The evaluation device (12) is designed to determine a higher confidence level for the object (8) if the object (8) is contained in the sensor data (6) of at least two monitoring units (2) generated at the same time and the monitoring fields of at least two monitoring units (2) at least partially overlap.

7. The security system network (1) according to claim 5 or 6, characterized in that The evaluation device (12) is designed to determine the trust level for the object (8) as a function of the quality of the sensor data (6) and / or the quality of the object information.

8. The security system network (1) according to claim 7, characterized in that a) the quality of the sensor data (6) depends on the physical characteristics of each of the monitoring units (2); and / or b) The quality of the object information depends on the position, movement direction and / or object category.

9. The security system network (1) according to claim 8, characterized in that The physical characteristics include the lifespan, type, error rate, failure rate, scattering rate, measurement method, installation location and / or trust information of each monitoring unit (2).

10. The security system network (1) according to claim 8, characterized in that The object class indicates whether it is a person (8a) or an autonomously driven vehicle (4).

11. The security system network (1) according to claim 5, characterized in that The evaluation device (12) comprises an AI module (13), wherein the AI ​​module (13) is designed to determine the trust level for the object (8) based on the sensor data (6) and / or the object information.

12. The security system network (1) according to claim 5, characterized in that The safety system network (1) comprises at least one autonomously driven vehicle (4), wherein at least one autonomously driven vehicle (4) is designed to receive the object list (7) from the central processing device (3).

13. The security system network (1) according to claim 12, characterized in that At least one of the autonomously driven vehicles (4) is designed to: a) if the objects (8) in the object list (7) whose distance from the intersection area (14) is less than a distance threshold have the trust level greater than a first threshold, at least one of the autonomously driven vehicles (4) enters the intersection area (14) without braking; and b) If the objects (8) in the object list (7) whose distance from the intersection area (14) is less than a distance threshold have the trust level less than a second threshold, at least one of the autonomously driven vehicles (4) enters the intersection area (14) at a reduced speed or stops.

14. The security system network (1) according to claim 13, characterized in that At least one of the autonomously driven vehicles (4) is designed so that, in the event of entering the intersection area (14) without braking, at least one or all safety systems of the autonomously driven vehicle (4) are bypassed or deactivated.

15. The security system network (1) according to any one of claims 12 to 14, characterized in that At least one of the autonomously driven vehicles is designed to travel in a truck (15) and / or a rail vehicle to load or unload cargo, wherein the autonomously driven vehicle (4) only drives into the truck (15) and / or the rail vehicle when it is within a distance range smaller than a distance threshold around the truck (15) and / or the rail vehicle and the objects (8) in the object list (7) meet the following conditions: a) the object (8) in the object list (7) is not a person; b) the corresponding object information of the object (8) has a trust level greater than a threshold value, The autonomous vehicle (4) is designed to bypass or deactivate one or all safety systems when driving into the truck (15) and / or rail vehicle.

16. The security system network (1) according to any one of claims 12 to 14, characterized in that At least one of the autonomously driven vehicles (4) is designed to continue driving even if at least one safety system for monitoring the environment fails, if the object (8) on the object list (15) will not cause a collision and the confidence level of the object (8) is greater than a threshold value.

17. The security system network (1) according to claim 1, characterized in that The area (5) is a warehouse or a factory.

18. A method for monitoring an area (5) in which objects (8) including autonomously driven vehicles (4) and persons (8a) move together, the method comprising the following method steps: - receiving (S1) sensor data (6) from a plurality of monitoring units (2), wherein the sensor data (6) contain objects (8) detected by the monitoring units (2) in the monitored area (5); - merging (S2) the received sensor data (6); - creating (S3) an object list (7) from said merged sensor data (6), wherein, The object list (7) contains the detected objects (8) together with corresponding object information; - transmitting (S4) the object list (7) to the autonomously driven vehicle (4).

19. The method according to claim 18, characterized in that The area (5) is a warehouse or a factory.

Images