Control network for mobile robots in a facility

Abstract

A control network (100) for supporting a plurality of mobile robots (130) operable in a facility (190) with stationary access points (140) configured to operate a basic RAN (145) compliant with a first radio access technology, RAT. The control network comprises: a network management system (120) to configure and perform network resource allocation in the basic RAN; and a fleet management system (110) to perform path planning and path execution for the mobile robots, for thereby carrying robot missions. The mobile robots includes one or more hotspot-enabled mobile robots (130*), each of which is equipped with a mobile AP (134) configured to operate a supplementary RAN (135) compliant with said first RAT, for thereby relaying network traffic between other mobile robots and the basic RAN. The fleet management system is configured to perform the path planning and path execution in view of locations of the hotspot-enabled mobile robots.

Description

CONTROL NETWORK FOR MOBILE ROBOTS IN A FACILITY TECHNICAL FIELD

[0001] The present disclosure relates to the field of mobile robot control. Inparticular, it discloses a control network for supporting mobile robots which operate in a facility with a basic radio access network (RAN), wherein some of the mobile robots carry a mobile access point operable to provide a supplementary RAN. Also disclosed is a method of operating such a control network. BACKGROUND

[0002] The technical context of the present disclosure resembles that of theapplicant’s earlier WO2022214193A1, which discloses a control network comprising a Fleet Management System (FMS) and a Network Management System (NMS). The FMS and NMS have separate responsibilities but coordinate and exchangeinformation in a well-defined way. The NMS manages a RAN which provides a set ofstationary access points (APs) with cellular or non-cellular radio coverage in at leastpart of a facility. The facility may for example be a factory, warehouse, port orcontainer terminal. The FMS is configured to plan the routes of a fleet of mobilerobots operating in the facility and may do so on the basis of a quality-of-service(QoS) forecast for said RAN. Conversely, the FMS may request a certain QoS levelfrom the NMS before it instructs the mobile robots to execute the planned paths.

[0003] It is common practice to plan the RAN – before commissioning – in a wayto accommodate communication requirements of the customers and applications that will be using this network. The planning may involve considering the envisioned number of devices and their area of operation. To extend communication range of a wireless network, multiple stationary points of attachment commonly need to bedeployed, e.g., Wi-Fi™ access points or 4G / 5G base stations. Adding more and morededicated access points or base stations is not always feasible, nor is the costjustifiable.

[0004] Furthermore, in many cases stationary APs are unable to address theproblem of intermittent so-called radio coverage holes. These ‘holes’ are areas withlow or insufficient signal quality, which can stem from reflectors or other obstacles to signal propagation which, from time to time, block the usable propagation pathsbetween an autonomous mobile robot and an AP of the RAN. Such reflectors andobstacles may include structural elements of a factory or plant, moving equipment, and even human personnel who carry metallic objects. If a mobile robot experiences poor wireless connectivity or loses it, fleet management system may become unableto control the robot or even track its position. This will, in turn, disturb the mobilerobot’s operation or may even lead to it being halted.

[0005] The state of the art includes various proposals for improving the RAN’scoverage locally, ultimately to prevent the occurrence of such coverage holes. Forexample, it is proposed in EP3669612A1 to equip each mobile robot in the facilitysuch that it can act as a mobile hotspot when needed. That document discloses anautonomous vehicle that carries a multifunctional radio module in addition to aconventional vehicular platform and motion processor. The radio module is operableto communicate with a fixed LTE network or another fixed radio network for backhauling purposes, and to provide signal to users in the autonomous vehicle’s area of coverage while the autonomous vehicle acts as a base station, and further for peer- to-peer communication with other autonomous vehicles for purposes of backhaul or coordination. Other disclosures relate to vehicles which have one wireless interface adapted for communication with a basic network using a first radio access technology (e.g., cellular 5G) and another wireless interface that allows it to communicate with other vehicles using a different, second radio access technology (e.g., Wi-Fi™).

[0006] To mention a further example, US20170164423A1 discloses a fleet ofautonomous vehicles (AVs), which are centrally coordinated by a backend system to perform delivery and other tasks. The routes to be traveled by the AVs may be optimized with respect to communication requirements, available networks and so forth. In cases where the AV has freedom how to solve an assigned transport task, the AV may utilize an up-to-date network map to plan its own optimal route. In fact, thecellular coverage in the area where the AVs operate is not only varying but sometimesinsufficient / absent; to address this, each AV in the fleet is equipped to set up a local mesh network for relaying communications between a cellular base station and a nearby AV that is too far from the base station, in a network-limited area. To facilitate such relaying, the backend system may transmit a dedicated network configuration for establishing this mesh network and it can also send out or reroute AVs to assist an AV that has driven out of network coverage.

[0007] In addition to the fully hotspot-enabled robot fleets and fully conventionalrobot fleet reviewed above, it may be expected that mixed fleets – where some mobilerobots are hotspot-enabled and some are conventional – will be operated in theforeseeable future. From a financial point of view, it is highly interesting to upgrade existing mobile robot systems by adding hotspot-enabled mobile robots without the need to modify the existing robots. Protecting the investments into so-calledbrownfield equipment in this way has the potential to save considerable effort andcost. The problem of controlling mixed robot fleets has not been adequately analyzedand solved in the prior art, however. To solve a given robot task, this control may include a combination of robot path generation and decision-making on how thelimited number of hotspot-enabled robots shall be utilized. These matters will beaddressed in the present disclosure. SUMMARY

[0008] One objective of the present disclosure is to make available a controlnetwork for supporting a plurality of mobile robots in a facility with stationary access points, under an assumption that some of the mobile robots are equipped with mobile access points (hotspot-enabled mobile robots). Another objective is to provide a method for controlling a fleet of mobile robots, from which some though not all arehotspot-enabled. A particular objective is to perform path planning and pathexecution in view of the locations of the hotspot-enabled mobile robots. Another particular objective is to optimize the use of a limited number of hotspot-enabled mobile robots in a fleet. Another particular objective is to propose a systematic path- planning approach for solving a given robot task by means of a mixed robot fleet.

[0009] At least some of these objectives are solved by the invention as defined bythe independent claims. The dependent claims relate to advantageous embodiments.

[0010] In a first aspect of the present disclosure, there is provided controlnetwork for supporting a plurality of mobile robots operable in a facility, in which one or more stationary access points are arranged. The one or more stationary access points are configured to operate at least one basic radio access network (RAN) compliant with a first radio access technology (RAT). The control network comprises: a network management system (NMS) with authority to configure and perform network resource allocation in said at least one basic RAN; and a fleet managementsystem (FMS) with authority to perform path planning and path execution for the mobile robots, for thereby carrying out one or more robot missions. The respective authorities of the FMS and the NMS are mutually exclusive. According to the first aspect, said plurality of mobile robots includes one or more hotspot-enabled mobile robots, each of which is equipped with a mobile access point configured to operate a supplementary RAN compliant with said first RAT, for thereby relaying network traffic between other mobile robots and the basic RAN. Further, the FMS is configured to perform the path planning and / or path execution in view of locations of the hotspot-enabled mobile robots.

[0011] The control network according to the first aspect may be said to addressthe connectivity problem at its core, namely, by taking the locations of the hotspot-enabled mobile robots into account already at the path planning stage or pathexecution stage, or both of these.

[0012] According to some embodiments herein, the locations of the hotspot-enabled mobile robots are considered by the use of a connectivity map of the facility, which indicates a quality of service (QoS) provided by not only the basic RAN but also by the supplementary RANs operated by the hotspot-enabled mobile robots. The path planning and / or path execution is performed in such manner as to avoid regions of the facility for which the connectivity map indicates an insufficient QoS.

[0013] In other embodiments, the locations of the hotspot-enabled mobile robotsare considered by performing the path planning as a combination of a path- generating process and an individualization process. The individualization process includes decision-making as to which paths are to be executed by hotspot-enabled mobile robots and which paths can be executed by mobile robots that are not hotspot- enabled.

[0014] In some embodiments, the FMS is configured to provide each plannedpath with an associated network traffic profile. A network traffic profile in this sensemay include a proposed RAN to be used by a mobile robot while it executes theassociated planned path, wherein the proposed RAN is selected from the basicRAN(s) and the supplementary RANs. This may enable the FMS to distribute the loadon the basic and supplementary RANs. The selection of the proposed RAN to each planned path may also serve as a verification that each path has sufficient connectivity to be executed.

[0015] Alternatively, the network traffic profile may indicate a proposedstationary AP from among the stationary APs in the basic RAN, wherein this AP is tobe used by a mobile robot during the execution of the associated planned path. This allows the FMS to assist the NMS by facilitating the NMS’s task of load balancingamong different APs; ultimately this could improve the distribution of the availableradio resources on system level.

[0016] Further alternatively, the network traffic profile may indicate parametervalues of a communication protocol to be used by a mobile robot during the executionof the associated planned path. This affords the FMS further control over thecommunications exchanged in the different RANs.

[0017] In some embodiments, the NMS has authority to configure and performtraffic management in the supplementary RANs. This extends the authority of the NMS compared to the state of the art, where the NMS manages the basic RAN. The traffic management may for example include setting a relative prioritization of, on the one hand, network traffic pertaining to a communication protocol for path executionand location tracking of the mobile robots and of, on the other hand, network trafficoriginating from or destined for independent software processes executing in the mobile robots. The second type of network traffic (e.g., data analytics) may in particular be destined for or originate from communication parties with which the software processes communicate over the global Internet. The first type of network traffic pertaining to path execution and location tracking generally need not be exchanged over the global Internet. Generally speaking, the system owner will prefer to minimize the second type of network traffic, or at least assign a significantly power priority to it than the first time of network traffic. The traffic management may further include enabling or disabling the mobile access point of a hotspot-enabled mobile robot. The disabling may be full or partial; partial disabling may include disabling the hotspot-enabled mobile robot from serving connecting clients other than the mobile robots, such as personal computers or mobile telephones. These traffic management options may help the NMS to ensure the smooth and efficient operation of the supplementary RANs at runtime.

[0018] In a second aspect of the present disclosure, there is provided a methodimplemented in a control network for supporting a plurality of mobile robotsoperable in a facility in which one or more stationary access points are arranged. Themethod comprises: operating at least one basic RAN compliant with a first RAT using the one or more stationary access points; identifying hotspot-enabled mobile robots among said mobile robots, each equipped with a mobile AP; operating supplementary RANs compliant with said first RAT using the mobile access points of the hotspot- enabled mobile robots, for thereby relaying network traffic between other mobile robots and the basic RAN; and performing path planning and path execution for the mobile robots, for thereby carrying out one or more robot missions, wherein the path planning and path execution are performed in view of locations of the hotspot- enabled mobile robots.

[0019] The present disclosure further relates to a computer program containinginstructions for causing a computer, or the control network in particular, to carry out the above method. The computer program may be stored or distributed on a data carrier. As used herein, a “data carrier” may be a transitory data carrier, such as modulated electromagnetic or optical waves, or a non-transitory data carrier. Non- transitory data carriers include volatile and non-volatile memories, such as permanent and non-permanent storage media of magnetic, optical or solid-state type. Still within the scope of “data carrier”, such memories may be fixedly mounted or portable.

[0020] Generally, all terms used in the claims are to be interpreted according totheir ordinary meaning in the technical field, unless explicitly defined otherwise herein. All references to “a / an / the element, apparatus, component, means, step, etc.” are to be interpreted openly as referring to at least one instance of the element,apparatus, component, means, step, etc., unless explicitly stated otherwise. The stepsof any method disclosed herein do not have to be performed in the exact order disclosed, unless explicitly stated. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Aspects and embodiments are now described, by way of example, withreference to the accompanying drawings, on which: figure 1 is an overview of a control network deployed in a facility where a plurality of mobile robots operate;figure 2 is a flowchart of a method implemented in a control network that supports a plurality of mobile robots operable in a facility in which a set of stationary access points is arranged; figure 3 shows an example functional structure of the fleet management system and the network management system; figure 4 shows an example hotspot-enabled mobile robot; figure 5 illustrates communication links with a mobile robot as one endpoint; and figure 6 is a sequence diagram illustrating operation of a control network according to embodiments herein. DETAILED DESCRIPTION

[0022] The aspects of the present disclosure will now be described more fullyhereinafter with reference to the accompanying drawings, on which certainembodiments of the invention are shown. These aspects may, however, be embodiedin many different forms and should not be construed as limiting; rather, theseembodiments are provided by way of example so that this disclosure will be thoroughand complete, and to fully convey the scope of all aspects of the invention to thoseskilled in the art. Like numbers refer to like elements throughout the description. System overview

[0023] Figure 1 is an overview in block diagram form of a control networkdeployed in a facility 190 where a plurality of mobile robots (MRs) 130, 130* operates. The facility 190 may be a building, factory, plant, warehouse, (partially outdoor) industrial environment, mine or the like. In some embodiments, the public road network is not a “facility” in this sense. Example MRs 130, 130* include self-propelled robots with a full-body motion capability, including automated guidedvehicles (AGVs), autonomous mobile robots (AMRs) and mobile manipulators (conceptually, they are mobile robots with one or more robot arms). The MRs 130, 130* may in particular be configured as industrial robots or other utility robots, such as healthcare robots.

[0024] In this example, a plurality of stationary access points 140 are arranged thefacility 190, and they are configured to operate at least one basic radio access network(RAN) 145. One basic RAN 145 may be operated by one stationary access point 140;or one basic RAN 145 may be operated by a group of stationary access points 140; orall stationary access points 140 may be configured to operate one common basic RAN145. The basic RAN 145 is compliant with a first radio access technology (RAT), suchas a cellular RAT, a non-cellular RAT, 3GPP New Radio (NR, or 5G), 3GPP Long Term Evolution (LTE), IEEE 802.11n / ac / ax / be (Wi-Fi™), or WIA-FA (WirelessNetworks for Industrial Automation – Factory Automation, specified in IEC PAS62948 and others). Correspondingly, a stationary access point 140 can be the access point of Wi-Fi™, the access device of WIA-FA, or a base station (NB, eNB, gNB) ofthe cellular network. If there are two basic RANs 145, these may for instance bedistinguished by carrying different service set identifiers (SSIDs) or access pointnames (APNs). The multiple basic RANs 145 may be spatially overlapping, partially overlapping, or disjoint.

[0025] As shown in figure 4 by way of example, an MR 130, 130* includes avehicular platform 132 allowing it to move two-dimensionally on a surface or three- dimensionally in air or a liquid. The vehicular platform 132 may comprise one or more of motion actuators, steering arrangements, wheels, tracks, impellers, moveablefins, etc. The MR 130, 130* further includes a robot manipulator 131 (with anoptional end effector), processing circuitry 133, a data memory 138 and a wirelessinterface (not shown). The wireless interface can be the (mobile) station of Wi-Fi™,the field device of WIA-FA, or a user equipment (UE) of the cellular network. Theprocessing circuitry 133 can be used for running general software 136 and / or dedicated software 137 relating to path execution and location tracking of the MR 130. The second type of software processes 137 can exchange sensor signals and control signals with sensors and actuators within the vehicular platform 132.

[0026] Figure 4 further shows an optional component of an MR, namely, a mobileAP 134. In the present disclosure, those MRs which carry a mobile AP 134 will be referred to as hotspot-enabled MRs (HEMRs) 130*. The HEMRs 130* are capable of operating a RAN of its own or operating a RAN together with further hotspot-enabledmobile robots (HEMRs) 130*. A RAN operated by at least one HEMR 130* will bereferred to as a supplementary RAN 135 in the present disclosure. The mobile AP 134 allows the HEMR 130* to operate a supplementary RAN for thereby serving one or more connecting clients with connectivity. The HEMR 130* remains capable of communicating as a client with a basic or supplementary RAN thanks to thepreexisting wireless interface. In the example of Wi-Fi™, the HEMR 130* may be equipped with dual radio hardware and dual Wi-Fi Medium Access Control (MAC)protocol endpoints, where one acts as client and the other as an AP. (Alternatively, inimplementations where the mobile AP replaces the wireless interface, this capabilityof communicating as a client may be contained in the mobile AP.) To summarize,while the wireless interface of an ordinary MR 130 is not configured to operate a supplementary RAN for the benefit of a connecting client, the mobile AP 134 enablesthe HEMR 130* to operate as an access point, access device, or base station.

[0027] In the present disclosure, the term mobile robot, or MR for short, shallinclude both ordinary MRs and hotspot-enabled MRs.

[0028] Returning to figure 1, the control network includes a network managementsystem (NMS) 120 with authority to configure and perform network resourceallocation in said at least one basic RAN 145. Preferably, the NMS’s 120 authority tocarry out such configuration and network resource allocation is exclusive at runtime. In the illustrative embodiment of figure 1, the NMS 120 communicates with the stationary access points 140 over a wired core network 141 that is also used for backhauling functions. The dashed circles around the stationary access points 140 indicate approximate coverage areas.

[0029] The control network further includes a fleet management system (FMS)110 which has authority to perform path planning and path execution for the MRs 130, 130*. The FMS 110 may have the exclusive authority to perform path planning and path execution during runtime. The respective authorities of the FMS and the NMS are mutually exclusive. The FMS 110 or some section thereof may execute Robot Operating System 1 or 2 (ROS1, ROS2).

[0030] The FMS’s 110 path planning and path execution may in particular includetask planning and task execution, wherein utility tasks are assigned to one or more of the MRs 130, 130*. Such tasks may be elements of higher-level robot missionsspecified by an operator or a system owner, or a project plan from a manufacturingexecution system (not shown) which the FMS 110 is to carry out. An example robotmission ^ may take the form of an ordered set of robot tasks ^^ which are disjoint or(partially) overlapping in time: ^ = (^^, ^^, … , ^^), wherein example robot tasksinclude navigation, workpiece manipulation, and full-body motion. Alternatively, therobot mission ^ is expressed as a utility-oriented result to be achieved, such astransportation mission, material-handling mission, and a desired final workpiececondition. The second option generally leaves the FMS 110 some freedom to decideon the exact way of fulfilling the robot mission ^. The path planning by the FMS 110 may support the assigned tasks or it may be non-productive, e.g., related to maintenance or parking of the MRs 130, 130*. The path execution may include providing movement commands (motion references) to the MRs 130, 130* wirelessly over the basic RAN 145 or another RAN.

[0031] The described functionalities of the FMS 110 and the NMS 120 can beachieved by a variety of software and / or hardware. Figure 3 shows a control network100 in which the FMS 110 and the NMS 120 each have a suitable example inner structure. The depicted inner structures are advantageous for mutual cooperation.What is visible in figure 3 primarily reflects the functioning of the FMS 110 and theNMS 120. The depicted components may correspond to an arrangement of physical components or an arrangement of portions of executable software code. The connectors in figure 3 represent typical flows of information occurring when the FMS 110 and the NMS 120 operate according to the present disclosure. The layout of these information flows does not preclude that information may travel along other paths in the internal structure, as is the case, e.g., when multiple components are connected to a common network with a star or mesh topology.

[0032] In the NMS 120, a network monitor 122 collects status information fromthe stationary APs 140 in the basic RAN(s) 145 and from the mobile APs 134 of the HEMRs 130*. The network monitor 122 may store the collected status information in a network database 123. In many cases, even if the stationary APs 140 are deployed and owned by the same entity as the MRs 130, 130* and facility 190, the FMS 110does not have direct access to low-level (or machine-level, or low-layer) statusinformation, which is the preferred basis for predicting network QoS. It is particularly common in newer wireless technologies to restrict access to such low-level statusinformation, including 3GPP 5G, WIA-FA, and Wi-Fi 6.

[0033] In the NMS 120, furthermore, a network configurator 121 configures thestationary APs 140 and the mobile APs 134 according to a resource allocation plan. The network configurator 121 may access the resource allocation plan in the network database 123.

[0034] A network resource allocator 127 in the NMS 120 generates the resourceallocation plan on the basis of the collected status information relating to the RANs 135, 145, which it can retrieve from the network database 123. The resource allocation plan may be generated, further, in view of maps of the facility which are supplied by a map database 124. The resource allocation plan may be generated, furthermore, in view of a predictive network resource request received from the FMS 110, as described in more detail below. The network resource allocator 127 stores the generated resource allocation plan in the network database 123.

[0035] In the NMS 120, there is further provided a network QoS forecaster 126.The QoS forecaster 126 is configured to predict the achievable network QoS based on the resource allocation plan and optionally based on map information relating to the facility.

[0036] Turning to the FMS 110, this system of the control network 100 isresponsible for path planning and path execution with regard to the MRs 130. The FMS 110 may have task planning and task execution as a further responsibility, wherein utility tasks (production, processing, material handling, transport etc.) are assigned to one or more of the MRs 130. Such tasks may be elements of a higher-level task or a project which the FMS 110 is to carry out. The path planning may support the assigned tasks or may be non-productive, e.g., related to maintenance or parking of the MRs 130. The path execution may include providing movement commands (motion references) to the mobile robots 130 wirelessly over the access networks 135, 145. The FMS 110 may have the exclusive authority to perform path planning and path execution during runtime. The FMS 110 or some section thereof may execute Robot Operating System 1 or 2 (ROS1, ROS2).

[0037] Within the FMS 110, there is provided a fleet monitor 112, which collectsstatus information from the MRs 130, 130* and may store it in a fleet database 113.

[0038] A path planner 117 generates the paths for the MRs 130, 130* and storesthem in the fleet database 113. The path planner 117 may be configured to generate the paths according to the collected status information of the MRs 130, 130*, which it may retrieve from the fleet database 113, and according to tasks assigned to the MRs130, 130* by the path planner 117. As mentioned above, the path planning may inparticular include task planning, wherein utility tasks (production, processing,material handling, transportation, etc.) are assigned to one or more of the MRs 130,130*. The path planning may further be based on maps of the facility, which are available from a map database 114. Alternatively or additionally, the path planning may further be based on an achievable (forecasted) QoS of the wireless network indicated by the QoS forecaster 126 in the NMS 120.

[0039] In the FMS 110, furthermore, a path executor 111 controls the MRs 130,130* so as to implement the paths from the fleet database 113 that were generated bythe path planner 117 (see above), also taking into account the status of the MRs 130,130* from the fleet database 113.

[0040] There is also provided a network demand forecaster 116, which is operableto predict a quantity, type and / or location of the required network resources. The network demand forecaster 116 may base this prediction on the MR paths determined by the path planner 117 and maps of the facility, the latter being available from the map database 114.

[0041] The FMS 110 may be described, in relation to at least one of the mobilerobots 130, as an edge computing resource. This is to say, processing circuitry in the FMS 110 is located, with respect to the topology of an access network 135, such thatthe mobile robot 130 is expected to enjoy a reasonable QoS in normal conditions, e.g.,the connection between the mobile robot 130 and the FMS 110 normally fulfils a minimum throughput, a maximum allowed latency, or similar requirements. Thefulfilment of such QoS requirements in normal conditions can be achieved by placingthe FMS 110 suitably in relation to the mobile robot 130 and / or configuring that access network’s 135 parameters related to routing, scheduling, resource allocation and traffic prioritization.

[0042] The communication from the FMS 110 to the NMS 120, and vice versa,proceeds over a communication channel 160. The communication channel 160 may be configured to support conventional unicast, multicast or broadcast messages. Alternatively, the communication channel 160 may be configured as a publication- subscription (or PubSub) service. The messages exchanged between the FMS 110 and the NMS 120 may include a predictive network resource request and its reply, a QoS request and / or a QoS report, a network traffic profile M1, an instruction M2 from the FMS 110 to the NMS 120 concerning configuration and traffic management in the supplementary RANs 135.Operating a control network to support a mixed robot fleet

[0043] The present disclosure relates to the situation where at least one of theMRs is a hotspot-enabled mobile robot (HEMR) 130* capable of operating asupplementary RAN 135. The FMS 110 of the control network is then configured toperform the path planning and path execution in view of locations of the hotspot- enabled mobile robots.

[0044] When the MR fleet includes at least one HEMR 130*, the control network100 can be advantageously operated in accordance with the method 200 depicted in flowchart form in figure 2.

[0045] In a first step 210 of the method, at least one basic RAN 145 compliantwith a first RAT (e.g., cellular, non-cellular, 3GPP 5G, 3GPP LTE, Wi-Fi™, WIA-FA) is operated using the stationary APs 140.

[0046] In a next step 211, the HEMRs 130* are identified among the MRs. EachHEMR 130* is equipped with an (active) mobile AP 134.

[0047] In a further step 212 of the method 200, a plurality of supplementaryRANs 135 compliant with said first RAT are operated using the mobile APs 134 of the HEMRs 130*. The supplementary RANs 135 are used, for example, for relayingnetwork traffic between ordinary MRs 130 and the basic RAN 145. Because theHEMRs 130* step in as relays, the range of the basic RAN 145 is effectively extended. The relaying chains may be single-hop or multi-hop chains. This is illustrated byfigure 1, where a two-hop relaying chain to the stationary AP 140-1 may includeHEMRs 130*-4 and 130*-7. This is likely to contribute to stronger, more stable and more reliable radio coverage from the viewpoint of MRs 130-5 and 130-6, which are located further from the stationary AP 140-1. This range extending effect may also benefit clients other than the MRs 130, such as personal computers, smart phones in the facility 190.

[0048] In a fourth step 213 of the method 200, path planning and / or pathexecution for the MRs 130, 130* are performed. An aim of the path planning and path execution may be to carry out one or more robot missions. The path planning and path execution are performed in view of locations of the HEMRs 130*.

[0049] In some embodiments of the method 200, within step 212, thesupplementary RANs 135 may be managed by the NMS 120, which configures andperforms traffic management in these networks. In some embodiments, the NMS 120carries out these duties in accordance with a configuration and traffic management instruction M2 that it receives from the FMS 110. The traffic management may further include resource allocation. In particular, with reference to figure 5, the traffic management by the NMS 120 may include performing a relative prioritization of -network traffic 520 pertaining to a communication protocol for pathexecution and location tracking of the MRs 130, and- network traffic 510 originating from or destined for independent softwareprocesses 136 executing in the mobile robots. The network traffic 520 of the first type may originate from or be destined for a path execution and location tracking software application 137, which in turn communicates with components in the vehicular platform 132. The FMS 110 may act as the other endpoint of this network traffic 520, which passes through the basic RAN 145 and / or supplementary RANs 135. The network traffic 510 of the second type may be destined for or originate from communication parties 502, which are not connected to any of the basic RAN 145 and supplementary RANs 135, but which are reached via the global Internet 501. (The category of network traffic 510 of the second type could also be considered to include traffic which neither pertains to path execution and location tracking, nor is exchanged over the global Internet, such ascommunication with other facility devices. Alternatively, this type of traffic is treatedas a third type, to which the NMS 120 is free to assign a separate priority level.) In some embodiments, the NMS 120 assigns a higher priority level to the first type of network traffic 520, so that productive and safe operation of the MRs 130 can continue even when network resources are limited. Indeed, the independent softwareprocesses 136 executing on the processing circuitry 133 of the MRs 130 are typicallyless critical from a system perspective.

[0050] In further embodiments of the method 200, the traffic management by theNMS 120 includes enabling or disabling the mobile AP 134 of an HEMR 130*. Thismay be done in the interest of prioritizing available resources, such as radio resourcesor energy resources. In particular, the FMS 110 may be configured to monitor abattery level and / or a processing headroom (amount of unused processing resources)of an HEMR 130*. If the FMS 110 determines that the battery level and / or aprocessing headroom is insufficient, it instructs the NMS 120 to disable the mobileAP 134 of the hotspot-enabled mobile robot. Optionally, the FMS 110 can order apartial disabling, such as disabling the mobile AP 134 in the HEMR 130* from servingconnecting clients which are not the MRs 130, or not MRs under the control of thecontrol network 100. A further form of partial disabling is to allow the HEMR 130* torelay just one of the types of network traffic 510, 520 discussed with reference tofigure 5 (or to relay just two out of three types of network traffic, if three types aredefined).

[0051] The present disclosure foresees a number of ways of performing, instep 213, the path planning and path execution in view of the locations of the HEMRs130*.

[0052] In one embodiment, the path planning and path execution are performedon the basis of a connectivity map of the facility 190 which the FMS 110 obtains. Theconnectivity map can be generated and maintained by the FMS 110 itself on the basisof readings of status parameters of the RAN(s) and / or radio measurements by fixedor robot-carried sensors. Alternatively, the FMS 110 can request the connectivity map from the NMS 120. The connectivity map can have the form of a data-enriched version of a map of the physical environment in the facility 190, e.g., a map which issuitable for conventional path planning and has been annotated with values of one ormore QoS metrics. The QoS metrics can be computed or predicted from the statusparameters and / or radio measurements. The values of the QoS metrics are local inthe sense that each value is associated with a point or a region of the facility 190; different points and regions of the facility 190 may be associated with different QoS metric values. The QoS metrics refer not only to the QoS provided by the basic RAN 145 but shall also take into account the contribution from any supplementary RAN(s) 135 which are operated in a given region of the facility 190. The computation or prediction of the QoS metrics can derive the location of a supplementary RAN 135from a planned path or tentative planned path of an HEMR 130*. The computation orprediction of the QoS metrics can take into account the expected load on the basicRAN 145 and supplementary RAN(s) 135 in a region, which can be derived from theexpected density of MRs 130 in this region in view of a planned path or tentative planned path of the MRs 130. With the connectivity map at hand, the FMS 110 is configured to perform the path planning and path execution in such manner as toavoid regions of the facility 190 for which the connectivity map indicates an insufficient QoS.

[0053] In another embodiment, the FMS 110 performs the path planning and pathexecution in view of the locations of the HEMRs 130*, and more precisely in suchmanner that (a) a local density of the HEMRs 130* is equalized over the facility 190,or (b) a local density of the HEMRs 130* is increased in regions of the facility 190 thathave an insufficient QoS (according to the connectivity map). If the FMS 110 uses anoptimization-based path-planning algorithm, then each of these options can be achieved by adding a reward term for spatially equal local HEMR density or a penaltyterm on spatially variable local HEMR density to the objective function of thealgorithm. The FMS 110 may also endeavor to meet both options (a) and (b), e.g., byincluding costs or penalties for both. If the FMS 110 uses another type of path-planning algorithm, the options can be achieved by modifying path-to-robot assignments, e.g., by letting a path intended for an MR 130 be executed instead by an HEMR 130*, or vice versa.

[0054] In another embodiment, the FMS 110 performs the path planning and pathexecution in view of the locations of the HEMRs 130* by means of a two-stage approach which comprises: -substep 213.1: a path-generating process which, on the basis of said one ormore robot missions, outputs a plurality of robot paths to be executed by respective mobile robots, which may be HEMRs 130* or ordinary MRs 130; and -substep 213.2: an individualization process which, on the basis of aconnectivity map of the facility 190 (see discussion above), outputs an indication that at least one of the robot paths is to be executed by an HEMR130*.

[0055] To illustrate the two-stage approach, the output of the path-generatingprocess 213.1 may be a data structure that includes the information in Table 1.Table 1: Example output of the path-generating process 213.1 Path ID Path description Assigned to robot IDP1 ^^(^): 0 ≤ ^ ≤ ^ MR1P2 ^^(^): 0 ≤ ^ ≤ ^ MR3P3 ^^(^): 0 ≤ ^ ≤ ^ MR5P4 ^^(^): 0 ≤ ^ ≤ ^ MR6P5 ^^(^): 0 ≤ ^ ≤ ^ MR8P6 ^^(^): 0 ≤ ^ ≤ ^ MR9The path description may have the form of a parameterized two-dimensional curve, which may be expressed as reference points in the facility 190 as a function of discreteor continuous time. It is possible to conclude from the path description – possibly byadditionally applying interpolation – the location of each assigned MR 130, 130* ateach time ^ in the range 0 ≤ ^ ≤ ^. Here, ^ denotes the duration of a time period forwhich the path planning is performed; in other words, the set0 ≤ ^ ≤ ^}includes all points visited by MR1. It is understood that the function ^^(^) is merely anominal description of the path, whereas an executing MR in a system may be allowed to deviate from it to a certain degree so as to improve the smoothness of thepath, to avoid unforeseen obstacles etc. In assigning the generated paths to individualMRs 130, 130*, the path-generating process 213.1 may take into account the level of equipment (e.g., means for lifting, welding, painting etc.), the operational status and other technical characteristics of the MRs 130, 130*, with the overreaching aim of carrying out one or more robot missions. As explained above, such information about the MRs 130, 130* can be read from the fleet database 113, which may have the example content shown in Table 2.Table 2: Example content of the fleet database 113 Robot ID Capability #1 Capability #2 Hotspot-Battery level enabled MR1 X 80%MR2 X X 25%MR3 X 80%MR4 X 80%MR5 X 80%MR6 80%MR7 X X 80%MR8 X 80%MR9 X 80%Still for purposes of illustration, it is assumed that the individualization process 213.2concludes – in view of the connectivity map – that the QoS is insufficient in someregions of the facility 190 visited by MR1 when carrying out path P1. Then, the output of the individualization process 213.2 is a modified version of the path assignment inTable 1, namely where MR1 has been replaced with an HEMR. To select a suitablesubstitute HEMR, the individualization process 213.2 respects the capabilities of MR1 and looks for a functionally equivalent HEMR in the fleet database 113. Therequirement for an HEMR with Capability #1 is met by MR2 and MR7, from whichMR7 appears to be the better choice in view of the battery level. The output of the individualization process 213.2 therefore has the appearance of Table 3.Table 3: Example output of the individualization process 213.2 Path ID Path description Assigned to robot IDP1 ^^(^): 0 ≤ ^ ≤ ^ MR7P2 ^^(^): 0 ≤ ^ ≤ ^ MR3P3 ^^(^): 0 ≤ ^ ≤ ^ MR5P4 ^^(^): 0 ≤ ^ ≤ ^ MR6P5 ^^(^): 0 ≤ ^ ≤ ^ MR8P6 ^^(^): 0 ≤ ^ ≤ ^ MR9

[0056] To illustrate another implementation of the two-stage approach, theoutput of the path-generating process 213.1 may be a data structure where the robot paths are not assigned to robot IDs but merely annotated with required capabilities ofthe executing MR. According to this alternative format, the example output data fromTable 1 has the appearance of Table 4: Table 4: Example output of the path-generating process 213.1 Path ID Path description Capability #1 Capability #2P1 ^^(^): 0 ≤ ^ ≤ ^ XP2 ^^(^): 0 ≤ ^ ≤ ^ XP3 ^^(^): 0 ≤ ^ ≤ ^ XP4 ^^(^): 0 ≤ ^ ≤ ^P5 ^^(^): 0 ≤ ^ ≤ ^ XP6 ^^(^): 0 ≤ ^ ≤ ^ XAgain, the individualization process 213.2 analyzes this output on the basis of the connectivity map. The output may be provided in the form of a further annotation of the robot paths, namely whether some of the paths shall be required to be executed by a hotspot-enabled robot, as illustrated by Table 5; the further annotated robot paths can then be used as a basis for assigning the paths to robot IDs. Table 5: Example output of the individualization process 213.2 Path ID Path description Capability #1 Capability #2 Hotspot-enabled P1 ^^(^): 0 ≤ ^ ≤ ^ X XP2 ^^(^): 0 ≤ ^ ≤ ^ XP3 ^^(^): 0 ≤ ^ ≤ ^ XP4 ^^(^): 0 ≤ ^ ≤ ^P5 ^^(^): 0 ≤ ^ ≤ ^ XP6 ^^(^): 0 ≤ ^ ≤ ^ XAlternatively, the output of the individualization process 213.2 is a final assignment of the robot paths to robot IDs referring to ordinary MRs 130 and HEMRs 130*. This will be identical to the information in above Table 3.

[0057] In implementations of the two-stage approach, the path-generatingprocess 213.1 can be carried out by any suitable path-planning algorithm, which is fedwith relevant inputs indicative of the robot mission ^ to be solved (see above). Inparticular implementations, the individualization process 213.2 includes balancing acost of using an HEMR 130* (e.g., added cost compared with using an ordinary MR130) against the benefit of having an improved QoS in those regions of the facility 190 that the HEMR 130* will visit. This may be achieved by maximizing a cost-benefit function with the following general appearance: ^(ℎ; ^) = ^^^^(ℎ; ^) − ^^^^^(ℎ), (1)where ℎ = (ℎ^, ℎ^, … , ℎ^) is a Boolean vector representing an HEMR-substitutiondecision for each of ^ robots (equivalent to the rightmost column ofTable 5 when ^ = 6),^ = {^^(^), ^^(^), … , ^^(^)} denotes the ^ robot paths generated in thepath-generating process 213.1, ^^^^(ℎ; ^) represents the benefit of the achieved QoS, and^^^^^ (ℎ) represents the cost of substituting HEMRs 130* for MRs 130according to the vector ℎ.The optimal substitution decision is given by ℎ∗ = argmax ^(ℎ; ^).^

[0058] Referring to the cost-benefit function (1), the benefit is preferably assessedfrom the point of view of the MRs 130 which execute the generated paths. The benefitcan be quantized as a multiple of a favorable QoS metric (e.g., Signal-to-Interference- plus-Noise Ratio [SINR], Reference Signal Received Power [RSRP], throughput) or as a multiple of an inverse of an unfavorable QoS metric (e.g., bit error rate, number of protocol data unit retransmissions, latency, delay). At least for large facilities 190, theadditional benefit from an HEMR 130*-i is included in the benefit term ^^^^(ℎ; ^) ona condition that the HEMR 130*-i and the benefitting MR 130-j are sufficiently closethroughout the planning period [0, ^]; for example, it may be assessed whether themaximum Euclidean distance between these robots stays below a threshold ^^throughout the planning period: ^ m^^a^x^^^^(^) − ^^(^)^ ≤ ^^.

[0059] To improve the accuracy of the individualization process 213.2, the cost-benefit function (1) can be weighted in various ways. For example, the benefit term can be weighted spatially based on the robot paths of the ordinary MRs 130:where ^(^) denotes a local density of the MRs 130. This is expected to lead to aprioritization of the HEMR substitution in the relatively more crowded regions of the facility 190. The weighting (2) can be further refined by also including information on the traffic demand by each MR 130, so as to reflect an increased ‘effective density’ in areas where the MRs 130 are expected to be more data-hungry.

[0060] In another example of weighting, the generated robot paths are partitionedinto paths related to collaborative robot tasks (or missions) and paths related tosingle-robot tasks (or missions): ^ = ^^^^^^^ ∪ ^^^^^^^ . Based on this partition, thebenefit term is separated into two differently weighted terms: ^(ℎ; ^) = ^^^^^^(ℎ; ^^^^^^^) + ^^^^^^^ℎ; ^^^^^^^^ − ^^^^^(ℎ), (3)where ^^, ^^ are constant scalar weights. If the weights are set such that ^^ > ^^ > 0,the HEMR substitution will be prioritized for the collaborative tasks. This is justified since the interruption of a collaborative robot task caused by a QoS insufficiency will lead to nonproductive downtime for multiple robots, whereas the interruption of a single-robot task is typically less consequential from an economic point of view.

[0061] Within the two-stage approach, the path-generating process 213.1 andindividualization process 213.2 may be performed as a sequence of steps which are separate in time. The sequence of steps can be iterated at least once, which can proceed as in the following example: 1. First execution of the path-generating process 213.1, in view of an initialconnectivity map indicating QoS provided by the basic RAN 145 only. 2. First execution of the individualization process 213.2, in which the outputfrom the first execution of the path-generating process 213.1 is modified such that some ordinary MRs 130 are replaced with HEMRs 130*. The modifications aim to improve QoS in regions of the facility 190 which have insufficient QoS. 3. Second execution of the path-generating process 213.1, in view of anupdated connectivity map indicating the QoS provided by the basic RAN 145 and the supplementary RANs 135 of the HEMRs 130*. 4. Second execution of the individualization process 213.2. Here, if the QoS isstill insufficient in some regions of the facility 190, the output from the second execution of the path-generating process 213.1 may be modified by replacing some further ordinary MRs 130 with HEMRs 130*. The iteration may be stopped when a satisfactory QoS has been reached throughout the facility 190, e.g., when the QoS is better than a threshold configured by the system owner or an operator.

[0062] Alternatively, the path-generating process 213.1 and individualizationprocess 213.2 may be performed as a common process.

[0063] In another embodiment, the FMS 110 performs the path planning and pathexecution (step 213) in view of the locations of the HEMRs 130* by generating – foreach planned path – an associated network traffic profile M1. The network trafficprofile M1 is added at least to those paths that are to be executed by ordinary MRs130. Conceptually, the network traffic profile M1 can be added as a further column in Table 1 or 4. In different implementations, the network traffic profile M1 indicates one or more of the following: -At least one proposed RAN, selected from the basic RAN(s) 145 and thesupplementary RAN(s) 135, to be used by an MR 130, 130* during the execution of the associated planned path. -At least one proposed stationary AP 140 within the basic RAN(s) 145 whichis to be used by a mobile robot during the execution of the associated planned path. -Parameter values of a communication protocol to be used by a mobile robotduring the execution of the associated planned path, such as parameter values of a communication protocol for path execution and location tracking. The parameter values may affect physical-layer aspects of thecommunication, or they could regulate higher-layer aspects such as a time resolution of the path execution. The parameter values may influence the robustness of the communication, the load on the RAN(s) or the corenetwork, or the usage of radio resources locally in a region of the facility 190 where the MR 130, 130* executing the path is present.

[0064] According to further embodiments, step 213 may include a substep wherethe FMS 110 generates, for a planned path, an associated predictive network resource request to be carried out by the NMS 120. The NMS 120 may handle the predictive network resource request by configuring or controlling the stationary APs 140 or the mobile APs 134 that the HEMRs 130* carry. Example

[0065] Figure 6 is an annotated sequence diagram which illustrates an exampleperiod of operation of a control network 100 (see figure 1) according to embodimentsherein. The vertical bars correspond to the following communication endpoints, which have been described in detail above: the FMS 110, ithMR 130-i, jthHEMR 130*-j, kthstationary AP 140-k, the NMS 120. The vertical dimension of the sequence diagram corresponds to time, where the events indicated relatively lower in the diagram shall be considered to occur later than the events indicated relatively higher.

[0066] It is seen in this example that the NMS 120 carries out the followingfunctions: -configure one or more wireless networks in all the Aps 140, 134, defining,among others, their unique network identifier and radio-specific parameters, -configure default network traffic routing and prioritization in one or moreAps 134, 140, also based on an MR’s 130, 130* traffic profile provided bythe FMS 110 and the operational status of each AP 134, 140,o as a minimum configuration, a highest network traffic priority maybe assigned to ensure that the FMS’s 110 tracking and control of thevehicles is supported,- update network traffic routing and prioritization in one or more Aps 134,140, in the events of, e.g., an ordinary MR 130 connecting to an HEMR130*, -monitor operational status of all the Aps 134, 140 as well as wirelessperformance of the MRs 130, 130*, and save them into the network database 123,- provide to the FMS 110 a generalized report on performance of the Aps 134,140 and the vehicles. Optionally, the NMS 120 may also: -specify for an MR 130, 130* a set of one or more preferred RANs for thevehicle to connect to, -configure other QoS support in an AP 134, 140, such as maximum allowedbitrate for an MR 130, 130* and / or one or more of its network traffic flows, -enable or disable the wireless interface or mobile AP 134 in an HEMR 130*which specifically allow other MRs 130 to connect to the HEMR 130*,- configure in an HEMR 130* a maximum number of other MRs 130 whichshall be allowed to connect to it as clients.The FMS 110, for its part, carries out the following functions in this example:- plan and control execution of missions and tasks for all the MRs 130, 130*,- plan and control execution of paths for all the MRs 130, 130*, also based ona map of the facility 190 with AP locations as well as the Aps’ operationalstatus provided by the NMS 120,o for an HEMR 130*, wireless performance is considered as well,- create a network traffic profile for an MR 130 (in view of its planned path)and send it to the NMS 120,- monitor operational status of all the MRs 130, 130* and save it into thefleet database 113.Optionally, the FMS 110 may also: -have the facility 190 divided into subareas, e.g. by indicating such subareason a map maintained by the FMS 110, which subareas reflect the number ofdeployed Aps but also include data on the Aps’ operational status (obtained from the NMS 120) and the wireless performance of the MRs 130, 130*(also obtained from the NMS 120),- plan robot paths so as to aim to distribute an equal number of the HEMRs130* across parts of the facility 190 which correspond to the map subareas, or to the parts in which ordinary MRs 130 and / or stationary Aps 140 more frequently experience a degraded communication performance, -update an MR’s 130 network traffic profile if, say, the MR 130 connects toan HEMR 130*, and send the updated network traffic profile to the NMS120, -request the NMS 120 to enable or disable a mobile AP 134 or a wirelessinterface of an HEMR 130*.Closing remarks

[0067] To summarize, the present disclosure covers fleet management takingadvantage of vehicles (in particular robots) which carry mobile APs. Planning pathsof such vehicles offers an adaptive approach to improving wireless communication range against occurrences of intermittent “radio coverage holes”; practical experience suggests that such holes occur relatively frequently.

[0068] The proposals in this disclosure can be implemented alongside existingvehicle deployments (so-called brownfield equipment). The proposals require minimal changes / upgrades or even none. In particular the mobile Aps (carried by HEMRs 130*) can match a basic level of the capabilities and features of their stationary AP counterparts, thus allowing the introduction of HEMRs 130* to betransparent to the existing vehicles on, e.g., the factory floor. Further, the mobile APscan be controlled, configured, and managed in a similar manner as their stationaryAP counterparts.

[0069] The present disclosure has further discussed the configuration of wirelessnetwork traffic routing and prioritization based on fleet management requirements. The proposals allow a differentiation as to how limited wireless network resources can be allocated for transmitting a mix of traffic from different applications and provide best possible performance for more critical network traffic (e.g., vehicle tracking and control).

[0070] The proposals herein may further enable a reduced power consumptionoverhead. More precisely, vehicles serving as mobile APs (i.e., HEMRs 130*) can experience an increased power consumption because of that role, which can be mitigated by the capability of the FMS 110 to request the associated AP functionalityto be disabled in HEMRs 130* whose battery level is low, wherein the APfunctionality is re-enabled later on a need basis or after charging.

[0071] The aspects of the present disclosure have mainly been described abovewith reference to a few embodiments. However, as is readily appreciated by a person skilled in the art, other embodiments than the ones disclosed above are equally possible within the scope of the invention, as defined by the appended patent claims.

Claims

CLAIMS1. A control network (100) for supporting a plurality of mobile robots (130)operable in a facility (190) in which one or more stationary access points (140) arearranged,wherein the one or more stationary access points are configured to operate at leastone basic radio access network, RAN (145) compliant with a first radio access technology, RAT, the control network comprising:a network management system, NMS (120) with authority to configure and performnetwork resource allocation in said at least one basic RAN; and a fleet management system, FMS (110) with authority to perform path planning andpath execution for the mobile robots, for thereby carrying out one or more robotmissions, wherein the respective authorities of the FMS and the NMS are mutually exclusive, characterized in that said plurality of mobile robots includes one or more hotspot-enabled mobilerobots (130*), each of which is equipped with a mobile access point (134) configuredto operate a supplementary RAN (135) compliant with said first RAT, for therebyrelaying network traffic between other mobile robots and said at least one basic RAN;and in that the FMS is configured to perform the path planning and path execution in view of locations of the hotspot-enabled mobile robots.

2. The control network (100) of claim 1, wherein said plurality of mobile robots(130) further includes mobile robots which are not hotspot-enabled.

3. The control network (100) of claim 1 or 2, wherein the FMS (110) is configured:to obtain a connectivity map of the facility (190), which indicates a quality of service,QoS, provided by said at least one basic RAN (145) and by the supplementary RANs(135); andto perform the path planning and path execution in such manner as to avoid regions of the facility for which the connectivity map indicates an insufficient QoS.

4. The control network (100) of claim 3, wherein:the FMS (110) is configured to maintain the connectivity map; or the NMS (120) is configured to maintain the connectivity map and share the connectivity map with the FMS.

5. The control network (100) of any of the preceding claims, wherein the FMS(110) is configured to perform the path planning and path execution for the hotspot-enabled mobile robots (130*) such that:a local density of the hotspot-enabled mobile robots is equalized over the facility(190); and / or a local density of the hotspot-enabled mobile robots is increased in regions of the facility that have an insufficient QoS.

6. The control network (100) of any of the preceding claims, wherein the FMS(110) is configured to perform: a path-generating process (213.1) which, on the basis of said one or more robotmissions, outputs a plurality of robot paths to be executed by respective mobilerobots; andan individualization process (213.2) which, on the basis of a connectivity map of thefacility (190), outputs an indication that at least one of the robot paths is to beexecuted by a hotspot-enabled mobile robot.

7. The control network (100) of claim 6, wherein:the path-generating process (213.1) outputs, for each robot path, an associated robot capability to be fulfilled by the executing mobile robot; and the individualization process (213.2) outputs an indication that the hotspot-enabled mobile robot which executes said at least one of the robot paths shall fulfil the robotcapability associated therewith.

8. The control network (100) of claim 6 or 7, wherein the individualization process(213.2) includes balancing a cost of using a hotspot-enabled mobile robot against a benefit of an improved QoS.

9. The control network (100) of claim 8, wherein the benefit of the improved QoSis weighted spatially based on the robot paths, in accordance with a local density ofmobile robots.

10. The control network (100) of claim 8, wherein the benefit of the improved QoSis weighted stronger for robot paths relating to collaborative robot missions than for generic robot paths.

11. The control network (100) of any of claims 6 to 10, wherein the path-generatingprocess (213.1) and the individualization process (213.2) are performed as a sequenceof steps separate in time.

12. The control network (100) of claim 11, wherein the sequence of steps is iteratedat least once.

13. The control network (100) of any of claims 6 to 10, wherein the path-generatingprocess (213.1) and the individualization process (213.2) are performed as a common process.

14. The control network (100) of any of the preceding claims, wherein the FMS(110) is configured to generate, for a planned path, an associated network trafficprofile (M1).

15. The control network (100) of claim 14, wherein the network traffic profileindicates one or more of the following: a proposed RAN, selected from said at least one basic RAN and the supplementaryRANs, to be used by a mobile robot (130) during the execution of the associatedplanned path; a proposed stationary access point (140) within said at least one basic RAN (145) tobe used by a mobile robot during the execution of the associated planned path;parameter values of a communication protocol to be used by a mobile robot duringthe execution of the associated planned path, such as parameter values of acommunication protocol for path execution and location tracking.

16. The control network (100) of any of the preceding claims, wherein the FMS(110) is configured to generate, for a planned path, an associated predictive network resource request to the NMS (120).

17. The control network (100) of any of the preceding claims, wherein the NMS(120) has authority to configure and perform traffic management in thesupplementary RANs (135).

18. The control network (100) of claim 17, wherein:the FMS (110) is configured to provide an instruction (M2) to the NMS (120) concerning the NMS’s configuration and traffic management in the supplementary RANs; and the NMS is configured to execute said instruction.

19. The control network (100) of claim 17 or 18, wherein the traffic management bythe NMS (120) includes relative prioritization of, -on the one hand, network traffic pertaining to a communication protocol forpath execution and location tracking of the mobile robots (130), and,- on the other hand, network traffic originating from or destined forindependent software processes (136) executing in the mobile robots.

20. The control network (100) of any of claims 17 to 19, wherein the trafficmanagement by the NMS (120) includes enabling or disabling the mobile access point (134) of a hotspot-enabled mobile robot (130*).

21. The control network (100) of claim 20, wherein the FMS (110) is configured:to monitor a battery level and / or a processing headroom of a hotspot-enabled mobile robot (130*); and in response to determining an insufficient battery level and / or a processing headroom, to instruct the NMS (120) to disable the mobile access point (134) of the hotspot-enabled mobile robot.

22. The control network (100) of any of claims 17 to 21, wherein the trafficmanagement by the NMS (120) includes enabling or disabling the mobile access point (134) of a hotspot-enabled mobile robot (130*) to serve connecting clients other than the mobile robots.

23. The control network (100) of any of the preceding claims, wherein said first RATis one of: a cellular RAT, a non-cellular RAT, 3GPP New Radio (NR), 3GPP Long Term Evolution (LTE), IEEE 802.11n / ac / ax / be.

24. A method (200) in a control network (100) for supporting a plurality of mobilerobots (130) operable in a facility (190) in which one or more stationary access points (140) are arranged, wherein the method comprises: operating (210) at least one basic radio access network, RAN (145) compliant with afirst radio access technology, RAT, using the one or more stationary access points;identifying (211) hotspot-enabled mobile robots (130*) among said mobile robots,each equipped with a mobile access point (134);operating (212) supplementary RANs (135) compliant with said first RAT using themobile access points of the hotspot-enabled mobile robots, for thereby relaying network traffic between other mobile robots and the basic RAN; and performing (213) path planning and path execution for the mobile robots, for therebycarrying out one or more robot missions, wherein the path planning and pathexecution are performed in view of locations of the hotspot-enabled mobile robots.

25. A computer program comprising instructions to cause the control network (100)of claim 1 to execute the method (200) of claim 24.

Images