WO2012131631A1

WO2012131631A1 - Automatically commissioning of devices of a networked control system

Info

Publication number: WO2012131631A1
Application number: PCT/IB2012/051547
Authority: WO
Inventors: Oscar Garcia Morchon; Petrus Desiderius Victor Van Der Stok
Original assignee: Koninklijke Philips Electronics N.V.
Priority date: 2011-03-31
Filing date: 2012-03-30
Publication date: 2012-10-04

Abstract

The invention relates to automatically commissioning of devices of a networked control system, particularly to automatically commissioning of light sources of a lighting system, and more specifically to assigning configuration parameters such as addresses related to locations of the devices in the network such as grid positions. An embodiment of the invention provides a method for automatically commissioning of devices of a networked control system, which comprises several devices, wherein each device is able to communicate with other devices via a communication means, wherein the commissioning comprises the following steps: - initiating a first phase for gaining local information in the system, wherein each device detects and identifies its closest neighbor devices and stores the closest neighbor devices (S10), and thereafter - initiating a second phase for gaining global information in the system, wherein one after another device assigns one or more configuration parameters, which are related to positions in the network, to one or more of its stored closest neighbor devices that have not yet assigned configuration parameters (S12). The assigned addresses and grid positions make it possible to calculate a multitude of predetermined light patterns with a set of commissioned light points.

Description

Automatically Commissioning Of Devices Of A Networked Control System

TECHNICAL FIELD

The invention relates to automatically commissioning of devices of a networked control system, particularly to automatically commissioning of light sources lighting system, and more specifically to assigning configuration parameters such as addresses related to locations of the devices in the network.

BACKGROUND ART

Networked control systems are a ubiquitous trend in commercial, industrial and institutional business markets and also in consumer markets. A typical example of a networked control system is a networked lighting system with dozens of networked, particularly interconnected light sources. In the future, it is expected that these networked lighting systems will evolve particularly due to new developments on lighting sources such as LED (Light Emitting Diode) luminaries leading to a higher number of light sources.

Networked lighting systems with a high number of light sources or luminaries are also reffered to as large scale networked lighting systems. These developments and needs are introducing, already today, several changes in the way of deploying, interacting with, and controlling the lighting systems used in a multiple of environments including offices, hotels, or home. As a networked lighting system will usually comprise a very high number of devices, traditional approaches used to install and interact with the system will become obsolete in the sense that the individual management of light sources becomes impractical or slow. Other methods are going to be needed to reduce installation cost and manage multiple lighting sources in a smart and unobtrusive manner without the burden of having to control each and every of the many lighting devices of the system.

The deployment, configuration, control, and management of such large-scale lighting systems are, however, complex.

First, due to the system scalability, an installer has to connect several tens or hundreds of luminaires that later should operate together. This is a prone to errors and costly task. An ideal solution to overcome these drawbacks would be a self-configuring system, which is able to automatically commission devices such as light sources and, thus, allows to save time and money during installation.

Second, the control of this new type of lighting systems is also complex. Having a higher number of luminaires allows, a priori, creating more advanced light settings. However, doing it is more challenging: For instance, if a light spot of a given size needs to be created at a given location, a number of luminaires should be involved so that each and every of those luminaires has to receive a message with the desired light settings. Therefore, an approach is desired that allows for such capabilities in a more efficient way.

WO2010/097737 A 1 relates to automatically commissioning of light sources of a networked lighting system, wherein commissioning messages are routed through a grid of light sources. The commissioning messages comprise hops counters, which may be updated on each hop of a message through the grid. Each light source has a location counter, which may be updated in accordance with the hops counter of a received commissioning message. This solution enables an automatic commissioning of devices of a networked control system with a minimum of technical overhead and in a fully automatic manner.

SUMMARY OF THE INVENTION

An object of the invention is to provide improved methods, systems, and devices for automatically commissioning of networked control systems, particularly large scale networked lighting systems.

The object is solved by the subject matter of the independent claims. Further embodiments are shown by the dependent claims.

A basic idea of the invention is to provide an algorithm for automatically commissioning of devices of a networked control system, which also assigns one or more configuration parameters, for example assigning as configuration parameters addresses to commissioned devices, which are related to locations of the devices in the network, particularly to locations in a grid of devices of the networked control system. Thus, known algorithms for commissioning of networked control systems with devices arranged in a grid can be improved in that a comfortable control of the devices of the system is provided since the device's addresses are related to their grid locations and, thus, enable an improved control. Instead of addresses, also other position related configuration parameters can be assigned to devices. For example, the commissioned devices can be configured to generate predetermined light patterns from the position of the light sources and the specified spatial light intensities. The inventive algorithm applies at least two steps for commissioning of devices of a networked control system and for assigning configuration parameters to the devices: in a first step for gaining local information in the system, each device or node detects and identifies its closest neighbor devices; and in a second step for gaining global information in the system, the local information gained in the first step is used to distribute and assign one or more configuration parameters to the devices, wherein the configuration parameters are related to the positions in the network.

An embodiment of the invention provides a method for automatically commissioning of devices of a networked control system, which comprises several devices, wherein each device is able to communicate with other devices via a communication means, wherein the commissioning comprises the following steps:

- initiating a first phase for gaining local information in the system, wherein each device detects and identifies its closest neighbor devices and stores the closest neighbor devices, and thereafter

- initiating a second phase for gaining global information in the system, wherein one after another device assigns one or more configuration parameters, which are related to positions in the network, to one or more of its stored closest neighbor devices that have not yet assigned configuration parameters.

The method may be implemented as an algorithm running in a distributed or a centralized fashion at different moments that allows allocating configuration parameters related to positions in the network, such as assigning addresses to lamps of a networked lighting system in an automatic fashion. The first phase of the method may be for example triggered by a commissioning tool, i.e. in a centralized manner, while the second phase may operate in a distributed manner or also in a centralized manner by means of for example the commissioning tool, which can locally run an algorithm that allows determining relative positions of each device in network.

The first phase may comprise the acts of

- transmitting a reference signal by the first device with its communication means, wherein the reference signal contains a device identifier of the first device,

- measuring a parameter such as the intensity, delay and/or time of flight of the reference signal received by second devices being neighbored to the first device,

- transmitting the measurement with a device identfier by each second device to the first device using the device identfier of the first device,

- sorting the received device identfiers of the second devices by the measurements, and - storing the sorted device identifiers of the second devices in the first device.

Thus, the detection and identification of closest neighbor devices can be particularly performed by means of intensity, delay and/or time of flight measurements, which are an indicator of the distance between the first device and a second device in the network. This method is for example advantageous if the devices do not have assigned addresses related to positions in the network, but usually only a global address such as a MAC address, which can be used as device identifier, which is sent back from a second device together with the measured intensity, delay and/or time of flight to the first device.

The communication means may comprise a light source and the transmitting of the reference signal may comprise broadcasting a coded light (CL) signal containing the device identifier of the first device, RF means and the transmitting of the reference signal may comprise broadcasting a RF signal containing the device identfier of the first device, and/or ultrasound means and the transmitting of the reference signal comprises broadcasting an ultrasound signal containing the device identfier of the first device.

In a networked control system with devices being arranged in a square grid the second phase may comprise the following steps:

a) selecting a device in the grid,

b) assigning as configuration parameter an address, which is related to an initial position in the grid, to the selected device,

c) assigning as configuration parameters addresses derived from the address of the selected device to at least one of its stored closest neighbor devices that have not yet an assigned address,

d) selecting a next device from the stored closest neighbor devices of the selected device,

e) repeating steps c) and d) until each of the stored closed neighbor device of the selected device has an assigned address.

In step d) a next device with the highest measured parameter, i.e. a maximum or minimum value among several measured parameters, for example the highest measured intensity, smallest delay or shortest time of flight, from the stored closest neighbor devices of the selected device may be selected.

In step c) addresses derived from the address of the selected device may be assigned to the first two devices of the stored closest neighbor devices if these devices have not yet an assigned address. In a networked control system with devices being arranged in a hexagonal grid the second phase may comprise the following steps:

a) selecting a device in the grid,

c) selecting a next device from the stored closest neighbor devices of the selected device and that does not yet have an assigned address,

d) assigning as configuration parameter an address derived from the address of the selected device to the selected next device,

e) assigning as configuration parameter an address derived from the addresses of both lastly selected devices to a further device that is stored as closest neighbor device in both lastly selected devices and has not yet an assigned address,

f) selecting the further device, and

g) repeating steps e) and f) until each of the devices of the grid has an assigned address.

A further embodiment of the invention provides a computer program enabling a processor to carry out the method according to the invention and as specified herein.

According to a further embodiment of the invention, a record carrier storing a computer program according to the invention may be provided, for example a CD-ROM, a DVD, a memory card, a diskette, internet memory device or a similar data carrier suitable to store the computer program for optical or electronic access.

Another embodiment of the invention provides a computer programmed to perform a method according to the invention and as described above.

A yet further embodiment of the invention provides a commissioning tool for a networked control system comprising

- communications means for communicating with devices of the networked control system, and

- processing means being configured to communicate a request to devices of the networked control system to perform a method of the invention and as described above.

A yet further embodiment of the invention relates to a device of a networked control system comprising

- communication means for communicating with other devices of the networked control system, - measuring means for measuring a parameter such as the intensities, delay and/or time of flight of reference signals received from other devices of the networked control system,

- storage means, and

- processing means being configured

- to sort device identfiers received from other devices of the networked control system together with measured parameters such as intensities, delay and/or time of flight of a broadcasted reference signal by the measured parameters and to store the sorted device identifiers together with measured parameters in the storage means,

- to perform a method of the invention and as described above.

A yet further embodiment of the invention relates to a system for automatically

commissioning of devices of a networked control system comprising

- several devices according to the invention, and

- a commissioning tool according to the invention for initiating a first phase for gaining local information in the system, wherein each device detects and identifies its closest neighbor devices and stores the closest neighbor devices, and thereafter initiating a second phase for gaining global information in the system, wherein one after another device assigns one or more configuration parameters, which are related to positions in the network, to one or more of its stored closest neighbor devices that have not yet assigned configuration parameters.

These and other aspects of the invention will be apparent from and elucidated with reference to the embodiments described hereinafter.

The invention will be described in more detail hereinafter with reference to exemplary embodiments. However, the invention is not limited to these exemplary embodiments.

BRIEF DESCRIPTION OF DRAWINGS

Fig. 1 shows an embodiment of a networked lighting system with a lighting controller and luminaries according to the invention;

Fig. 2 shows four configurations of luminaries in a square grid of a networked lighting system; Fig. 3 shows step 0 of an embodiment of an algorithm according to the

invention;

Fig. 4 shows further steps of an embodiment of an algorithm according to the invention for assigning addresses to luminaries in a square grid of a networked lighting system;

Fig. 5 shows several steps of an address assignment to luminaries in a square grid of a networked lighting system according to the invention;

Fig. 6 shows several steps of an address assignment to luminaries in a

hexagonal grid of a networked lighting system according to the invention;

Fig. 7 shows a flowchart of an embodiment of the method for automatically commissioning of lamps of a networked lighting system according to the invention;

Fig. 8 shows a flowchart of an embodiment of step S10 of the method for automatically commissioning of lamps of a networked lighting system according to the invention;

Fig. 9 shows a flowchart of an embodiment of step S12 of the method for automatically commissioning of lamps of a networked lighting system according to the invention, wherein the lamps are arranged in a square grid;

Fig. 10 shows a flowchart of an embodiment of step S12 of the method for automatically commissioning of lamps of a networked lighting system according to the invention, wherein the lamps are arranged in a hexagonal grid.

DESCRIPTION OF EMBODIMENTS

In the following, functionally similar or identical elements may have the same reference numerals. Furthermore, embodiments of the invention are described by means of networked lighting systems even if the present invention is generally applicable to any kind of networked control system comprising several devices with a given arrangement.

The embodiments are described for regular grid arrangements of devices even if the present invention is generally applicable to many other device arrangements such as circular distributions, semi-regular grids, etc. The term "lighting system" used in the following refers to a networked lighting system, i.e. a lighting system with a network for connecting all luminaries with one or more lighting controllers via a wired and/or wireless network.

The terms "luminary" and "lamp" refer to a light source with a network interface, i.e. an electronic circuitry for connecting the light source to a network and for receiving and transmitting data from other light sources or a lighting controller or a commissioning tool of the networked lighting system. A "luminary" or "Lamp" corresponds to a device of a networked control system.

The term "node" refers to a network node, which is individually addressable in the networked lighting system. A node may be for example implemented in a single luminary/lamp/light source or a lamp controller controlling several lamps connected to it. A node may be for example also a network device such as a router, switch or hub to which several controllable devices of the networked control system can be connected.

A luminary or lamp may comprise one or more light sources, for example several LEDs, which are controlled by the same control circuitry. A luminary must not contain control circuitry and light source(s) in a one unit, for example in the same housing, instead the control circuitry may be also arranged external and provided to control several light sources for example via a bus interface.

A luminary or lamp can especially be a smart LED retrofit lamp (sLrL). In a sLrL system, smart lamps will be outfitted with wireless interfaces and new control algorithms allowing them to discover, recognize, and cooperate with close-by sensor devices. One of the features of those systems is the capability of creating complex light settings or continuously adapting to external light contributions (sun light) in order to reduce the energy requirements.

Two of the main functionalities required in a networked lighting system are the identification of lamps and the design of efficient multicast to address a number of light sources. A sensor needs a lamp identifier, e.g., to unambiguously identify a lamp during system operation. Unique identification can be done by means of the MAC (Media Access Control) addresses, short MAC addresses as used for IEEE 802.15.4 or Internet addresses (Internet Protocol (IP) addresses). However these addresses do not relate to the location of the lamp.

Therefore, before operating a networked lighting system, all luminaries or lamps, which should be individually addressable, must be commissioned, i.e. introduced to control instances of the networked lighting system such as a lighting controller in order to allow to centrally control and generate a desired lighting environment. Furthermore, an addressing scheme must be applied to the system, which allows addressing luminaries individually.

In order to accomplish this, the present invention pursues an advanced approach relying on smartly assigning addresses during commissioning. In a first phase lamps are addressed by means of their unique MAC addresses. Afterwards, new (IP) addresses which encode the location enable protocols running to benefit from these new addresses. For instance, smart addressing schemes can bring advantages such as the capability of addressing and creating light patterns with a set of devices that might be in close vicinity or symmetric positions. In the following, it will be discussed how to generate and distribute these smart address according to the invention by means of embodiments. Related approaches rely on them for efficient multicast and broadcast. First, some relevant use cases are analyzed that are addressed by the invention. Then, embodiments of a number of algorithms for address assignment and efficient multicast of lighting control messages according to the invention enabling the described use cases are described.

The invention can be deployed with several models for lighting systems that differ in the application scope (indoor and outdoor lighting) or the way the luminaries are distributed.

• Uniform grid of luminaries for indoor lighting: the lighting system comprises a (square or hexagonal) grid of luminaries uniformly distributed in the ceiling following the X and Y axis. Each direction comprises a total of mx and my luminaries so that the total of lamps in the room is equal to M = mx * my. Each luminary has usually been assigned an identifier such as a MAC address.

• Hierarchical grid of luminaries for indoor lighting: the lighting system comprises an L level hierarchical grid of luminaries uniformly distributed in the ceiling following the X and Y axis. Each direction comprises a total of mx_l and my_l luminaries at level 1 with 1 < 1 < L. Thus, the total number of lamps at a given level is equal to ml = mx_l * my_l and in the room equal to M =∑ Ml.

• Line of luminaries for outdoor lighting: a set of luminaries, e.g., distributed along a line as in a street. Each luminary can be assigned its GPS position.

Use cases for the above deployment approaches assuming that each luminary has been assigned one of these smartly distributed identifiers: ·

• Control of several luminaries according to a given pattern in close location: An important use case refers to the creation of a personalized light spot at a given location. This is motivated by a user moving through the room with his desktop and the requirement of translating the light spot. A user can be located at (xu, yu) that uses a sensing device to configured the light spot. ·

• Generation of a lighting pattern in a room allows, e.g., adapting to external conditions or the presence of a number of users. It also allows creating a light pattern such as a gradient. In a room comprising M luminaries and being managed by a room controller, it requires the controller to send a unicast message to luminaries. ·

• Generation of a lighting pattern such as a sinusoidal, square, or triangular wave further extends the last use case. In this case, a lighting system can be used to create different light effects in a given space. This light pattern is configured by a controller or sensing device.

The invention can use different communication technologies in the lighting system, particularly coded-light (CL) so that the lamps can transmit an identifier in the light. Additionally, lamps of the lighting system can establish bidirectional communication links over CL or RF communication links between lights and sensors. Other communication technologies are also possible, either wired or wireless.

Assuming that luminaries are distributed according to a square grid, a distributed algorithm, that allows allocating addresses to the devices in an automatic fashion according to the invention, is described in the following in detail.

The invention is applied to a grid of luminaries. Typical grids are square grids or hexagonal grids.

In a square grid, there exist four possible configurations for luminaries, as shown in Fig. 2: a lamp is in configuration I if it is in a corner of the grid that has seven neighbors, a lamp is in configuration II if it is on a corner that has three neighbors, a lamp is in configuration III if it is in the middle of the grid that has eight neighbors, and a lamp is in configuration IV when it is on the border of the grid and, thus, has five neighbors.

In a normal square grid of luminaries, there exist zero luminaries in configuration I (however, square grids may exist with luminaries in configuration I), four luminaries with configuration II; 2(mx - 2 + my - 2) luminaries in configuration IV; and (mx - 2)(my - 2) nodes in configuration III.

It is assumed that each lamp is associated to a long unique identifier such as a MAC address. This MAC address is used to address a device during commissioning.

Each lamp can be outfitted with a sensor so that a lamp can obtain reliable light intensity measurements from its neighbors. Alternatively to the last assumption, each lamp can be outfitted with an ultrasound device and a lamp can use it to reliably determining the relative distance to its neighbors.

Each lamp has a communication interface. This interface might be based on CL or RF technology. If a CL interface is applied, this CL interface may also be used to measure the intensity of light signals broadcasted by neighbored lamps. If a RF interface is applied, this RF interface may also be used to perform delay and time of flight measurements of RF signals broadcasted by neighbored lamps.

Each lamp is also able to communicate with its directly neighbored lamps in the grid with the communication interface or an alternative communication interface, for example a ZigBee™, WiFi™, Bluetooth®, Ethernet, DALI, or an IP enabled network such as 6L0WPAN/C0RE interface.

In Fig. 1, an embodiment of a networked lighting system 10, which comprises a lighting controller 14 and a network of twenty luminaries 22 being arranged in a square grid, is shown. Such a lighting system can be for example installed in an office, a greenhouse, or a large hall. The lighting controller 14 can transmit a message to each of the luminaries 22, which may comprise control information for a luminary 22, addressed by the message. The control information may comprise for example a command for a desired light setting of the addressed luminary 22, for example to set its lighting intensity to a certain level or to change its lighting color, or a commissioning request, which sets the addressed luminary 22 into a special commissioning mode as will be described later. Thus, the lighting controller 14can also serve as a commissioning tool for the networked lighting system 10.

The lighting controller 14 , which may be implemented for example by a PC (Personal Computer) configured with a dedicated software for lighting control and

commissioning, comprises a memory 16 (for example a ROM, RAM, Flash, SSD, HDD, CD, DVD) storing a software and a processor 18 (for example a standard PC processor or a special microcontroller for lighting systems) configured by the software to control the lighting created by the luminaries 22. The lighting controller 14 further comprises a transceiver 20 for transmitting and receiving messages over the network to or from the luminaries 22. The transceiver 20 may be for example adapted to communicate according to a wired or wireless communication technology, particularly according to one or more of the following standards or technologies: ZigBee™, WiFi™, Bluetooth®, Ethernet, DALI, IP enabled network technologies such as 6L0WPAN/C0RE.

Each of the luminaries 22 is adapted to receive messages from the lighting controller and to set its lighting in accordance with a control information contained in a received message. For receiving the messages, a luminary 22 comprises a transceiver 24. A luminary 22 further comprises a memory 28 (for example a RAM, Flash) and a controller 26 (for example a microcontroller) configured by a program stored in the memory to process received messages and to set a lighting created by the luminary 22 in accordance with a control information contained in the processed message and obtained by the message processing.

Each of the luminaries 22 is also able to communicate via its transceiver 24 or an additional communication interface with other luminaries 22 in the grid or the lighting controller 14. For example, a luminary 22 can transmit its status to the lighting controller 14 or transmit a message to its directly neighbored luminary 22 for routing the message through the grid to another luminary 22. As already outlined above, communication can be for example performed wirelessly via CL and/or RF technology.

In the memory 28 of a luminary 22 its own address, measurements and identifiers of neighbored luminaries may be stored as will be described later in connection with the embdiments of algorithms according to the invention.

After commissioning and address assignment, each luminary 22 in the grid has an unique addressrelated to a location of the luminary 22 in the grid, as shown in Fig. 1. The relation of an address to a location can be implemented for example in the form of a table containing all assigned addresses and corresponding locations in the grid or a kind of IP address (if the network of the system 10 is an IP based network) with a location depending attachment. "Location" means the location in a certain coordinate system, which is known to the lighting controller 14 and can be used to efficiently address and control luminaries required to generate a certain lighting at a desired location. For example, the address of each luminary 22 of system 10 can be related to a logical ordering of the luminaries, as indicated in Fig. 1 by the square grid or matrix arrangement of the luminaries and the column y (0-4) and line x (0-4) indices: for example the luminaries 22 with configuration II (refer to Fig. 2) arranged in corners of the matrix have the logical positions (0, 0), (0, 4), (4, 0), and (4, 4), the luminaries 22 with configuration III (refer to Fig. 2) have the logical addresses (1, 1), (1, 2), (1, 3), (2, 1), (2, 2), (2, 2), (3, 1), (3, 2), (3, 3), the luminaries 22 with configuration IV (refer to Fig. 2) habe the logical addresses (0, 1), (0, 2), (0, 3), (1, 0), (2, 0), (3, 0), (4, 1), (4, 2), (4, 3), (1, 4), (2, 4), (3, 4). The logical address of a luminary defines a relative location in the grid.

Next, the inventive algorithm steps for commissioning and address assignment are explained in detail. Basically, the algorithm runs in three main phases, wherein the first phase is a general setup of the system immediately after installation and the second and third phase are the essential phases of the inventive algorithm. In the following, the term "node" is used as synonym for a network node, i.e. an individually addressable element of a networked lighting system. A node can be implemented by a certain network device, for example by a luminary with a network interface or a hub or router for controlling several luminaries. A node has a unique identifier, but is out of the box unconfigured for network usage and, thus, has to be configured for full usage in a networked lighting system.

The first phase of the algorithm, network up, allows a network of fully unconfigured nodes to get knowledge of the devices in the network, i.e., each device becomes aware of the MAC addresses of its neighbors. In a second phase, each node has to become aware of the relative positions of its neighbors. At this time, a node only knows which nodes are close and which are far. This can be done based on the amount of light received or a ultrasound intensity, or delay or TOF (Time of Flight) RF measurement. However, after the following first two phases the nodes are not yet able to distinguish between their relative positions. Therefore, the final phase aims at assigning global addresses from the local information known to each node.

PHASE 1: Network up - Getting started

To setup the network, it is assumed that the networked lighting system is being installed in a room, e.g., an office. In such a setting, an installer takes luminaries 22, containing the nodes, from a box and plugs them in the ceiling. After installation and powering up, each node can start a neighbor discovery protocol based on existing network protocols such as IEEE 802.15.4 and an IP neighbor discovery protocol. In IP networks, each device assigns itself a unique IP address, possibly generated from its MAC address.

If a 6LoWPan (acronym for IPv6 over Low power Wireless Personal Area Networks) border router exists in the networked lighting system, for example implemented in the lighting controller 14, the 6LoWPan border router knows the addresses of all connected nodes, as prescribed by the 6LoWPan neighbor discovery. An application in the 6LoWPan border router can then execute the following actions:

It discovers the nodes which are connected to luminaries. To this end, each luminary has to broadcast its unique address.

It sends to all connected luminaries a message "light ON".

Once the installer has visually checked that all the luminaries are ON' meaning that they are successfully registered by the 6LoWPan border router, the installer can use a commissioning tool, for example the lighting controller 14 or a dedicated tool such as a handheld tool, and press a button called 'GAIN GRID LOCATION INFORMATION'. This button triggers the second phase (PHASE 2) of the algorithm.

In addition to the above general setting, the following more specific approaches might apply:

The installer relies on a commissioning tool (alternatively to the above

6LoWPan border router). The installer can control the beginning of the automatic installation - he initiates the broadcasting of a 'START CONFIGURATION' broadcast message once he has rolled out the whole network. Upon reception of the 'START CONFIGURATION' message, each luminary starts a neighbor discovery protocol based on existing network protocols such as IEEE 802.15.4.

An application running on the 6LoWPan border router or the commissioning device can handle and avoid collisions between devices by setting a random time T. Any connected device has to reply to the message by broadcasting its unique address at random time t between 0 and T. After sending the broadcast message each luminary remains ON. Each luminary in the network as well as the commissioning tool records the broadcasted addresses. In the above description, the maximum time T can depend upon the network size, i.e., the number N of luminaries in the network.

In order to enhance the system robustness, the commissioning tool can resend the 'START CONFIGURATION' message several times. Luminaries that have already broadcasted their unique (IP) addresses remain silent, but those, which were not able to broadcast it, will do it now. Note that if some Luminaries are not 'ON' after a few trials, the installer can easily identify those devices because they will be 'OFF', check their installation, and resend the 'START CONFIGURATION' message again.

Once the installer has visually checked that all the luminaries are 'ON' meaning that they have successfully broadcasted their unique addresses, the installer can use his commissioning tool and press a button called 'GAIN LOCAL INFORMATION'. This button triggers the second phase of the algorithm (step S10 in Fig. 7). For example, the installer can press a button on a graphical user interface (GUI) of a lighting control program executed by the lighting controller 14 and displayed on a monitor. The pressing of the button activates a commissioning program or routine of the lighting control program for

commissioning the luminaries 22 of the system 10. The commissioning particularly comprises transmitting a broadcast "GAFN LOCAL FNFORMATION" via the transceiver 20 of the lighting controller 14 to the luminaries 22. Next, the second phase is described in detail. PHASE 2: Gaining Local Information

Gaining local information generally means that each lamp becomes aware of the devices that are closer or further away. Particularly, it means that each lamp detects and identifies its closest neighbor lamps in the grid, for example a lamp 22 at position (3, 3) in the lighting system of Fig. 1 detects and identifies alls neighbor lamps in its closest

neighborhood 12. Lamps detected and identified are stored internally in a lamps 's 22 storage 28 for a later usage in the third phase of the algorithm.

The gaining of local information can comprise the transmitting of a reference signal by a first lamp 22 via its transceiver 24, for example a RF or a light signal or an ultrasound signal (step S 100 in Fig. 8). The reference signal can be a broadcast signal and contains an identifier of the transmitting lamp, typically its MAC address, which uniquely identifies the first lamp in the networked lighting system 10. Each other lamp (second lamps) in the grid can receive the reference signal and measure the intensity, delay and/or time of flight of the received reference signal, for example by using a built-in sensor or by using a measurement of an external sensor located in the vicinity of the lamp and connected to the lamp (step S102 in Fig. 8). Each second lamp transmits the measurement back to the first lamp by using the identifier of the first lamp contained in the reference signal, such as the MAC address (step S104 in Fig. 8). The back-transmitted signal contains the measurement of the sending second lamp and the identifier of the measuring second lamp. The first lamp can control the reception area for example by limiting the time for receiving back-transmitted signals from neighbored devices. After receipt of the back-transmitted signals from the second lamps, the first lamp sorts the received second lamp identifiers by the measurement, for example from the highest measured intensity to the lowest one or the shortest delay or time of flight to the longest one (step S106 in Fig. 8). This sorting of the received second lamp identifiers together with their measurement values is stored in the internal memory 28 of the first lamp 22 (step S108 in Fig. 8). Since a measurement corresponds to the distance between the first lamp and a second lamp, the sorted list contains at its beginning the second lamps, which are closest to the first lamp, i.e. the closest neighbor second lamps in the grid.

In the following, some further details about implementations of the second phase with RF and CL technology as communication technology between the luminaries in the lighting system are described. The reference signal is transmitted as light signal.

RF Communication

The usage of RF communication means that it can be assumed that each luminary has received the broadcast messages of each other device in the network. This is a valid assumption in most of the cases because RF can easily reach a range of 20 - 50m. If 6LowPAN is applied to the lighting system, the LowPAN neighbor discovery algorithm can be used and it can be further assumed that all luminaries are known to a 6LowPAN border router (6LBR) because their IP addresses are stored. A commissioning application installed in the 6LBR can then first ask all the luminaries in the network to switch off. In case of a commissioning tool instead of a nor mal 6LBR, after pressing a ΌΑΓ LOCAL

F FORMATION' button, the commissioning tool can first ask all the luminaries in the network to switch off. Then the 6LBR or the commissioning tool starts requesting each luminary j in the network (the 6LBR or the commissioning tool are aware of the unique addresses of the luminaries in the network) to switch on for a time ΔΤ at a given reference intensity I ref . Light propagation depends upon factors such as the distance or radiation pattern of a given luminary, thus, for identical luminaries, more closely located luminaries will receive a higher light intensity. During this time, the rest of the luminaries must record the measured amount of light from lamp j. Each lamp jO must then send the measured value to lamp j. To avoid collisions, each lamp jO can send its value at time t between 0 and T.

Lamp j arranges the measurements in a well- arranged vector of light intensities VALj. Based on this information, Lamp j becomes aware of the light intensities, and thus distances, of different lamps jO.Another possibility for the above algorithm consists in exploiting symmetries to reduce the communication overhead. In other words, in some settings it can be assumed that if lamps j and jO shine at reference light intensity Iref, then they receive the same light intensity from each other. In this case, a lamp only has to store the received light intensities of other lamps to construct its vector of light intensities VALj. This removes the requirement of each lamp having to receive the amount of light that other luminaries have sensed. If lamp j is in configuration I (II or III or IV), then the seven (three, eight, or five) first elements of its vector VALj will be approximately equal and the rest smaller. The vector containing these elements can be denoted as the relevant well-arranged vector of light intensities VARLj that contains the direct neighbors of the lamp. Thus, this first step is used by a lamp to find out to which configuration it belongs. Each lamp jO might also send the sensed information to the commissioning device. ·

CL Communication

If CL is used, it cannot be assumed that all the nodes received a message in a single hop. The reason is that light propagation is, a priori, much weaker than RF. A way of addressing this issue is by using networking capabilities (routing) on top of a simple coded- light communication channel. However, this is not an actual requirement in many cases because at this stage only local information is of interest. In this case, an installer might press the 'GAIN LOCAL INFORMATION' button while pointing at one of the luminaries.

Pointing refers to the transmission of the message 'GAIN LOCAL INFORMATION' over CL to the target device. The commissioning device can address a given device because its address is transmitted in CL. Upon reception of this message, the target luminary j will shine at reference light intensity Iref while broadcasting a message 'GAIN LOCAL INFORMATION'. Each Luminary node jO receiving light (and thus the message 'GAIN LOCAL INFORMATION') from node j will (i) measure the about of light received from that luminary node and (ii) rebroadcast the message 'GAFN LOCAL INFORMATION' with its own address. By applying the same symmetry assumption described above under RF

Communication, each Luminary j only has to sense the intensity of the received message 'GAIN LOCAL INFORMATION' to construct its vector of well-arranged light intensities VALj .

Next, the third phase of the inventive algorithm is described in detail (step S12 in Fig. 7).

PHASE 3: Gaining Global Information

The last phase of the algorithm refers to the usage of the local information gained during the second phase to distribute unique addresses according to a grid, which relate to the positions of the lamps in the grid. In an embodiment of the invention, the following steps are performed:

1) First, an initial device is selected (step S120 in Fig. 9). Selection may be performed automatically or manually by means of a commission tool, for example the lighting controller 14. One way can be that the lighting controller 14 automatically selects an initial device after all luminaries 22 of the system signaled the receipt of measurements from second devices. Another way could be that an installer waits until the lighting controller 14 signals that all luminaries 22 have received their measurements from second devices

A selection can be performed for example in the following way: each of the four luminaries belonging to the configuration shown in Fig. 3can vote to choose a leader by broadcasting a random number k_j in [0, K - 1] together with its unique address. The node with the highest k_j is chosen as the leader. If two or more nodes choose the same value, those nodes repeat the voting until they agree on a leader. Alternatively, an installer might manually choose one of these four luminaries. This could be done by using coded light and pointing to one of the devices in one of the corners with a dedicated handheld tool provided to control luminaries 22 with CL signals. 2) The protocol for address assignment works in steps or rounds. Each node j has to keep track of: (a) the number rounds; (b) the devices that have received an address; and (c) the devices in its VARLJ that have not received an address yet.

3) In Step 0, the leader starts the address assignment process by choosing an initial address, for example address (0,0) in a square grid such as the one shown in Fig. 3 for itself (step S122 in Fig. 9) and address (1,0) and (0,1), which are derived from the leader address, for the two first devices in its vector VARL_Leader . These devices are denoted as node(l, 0) and node(0, 1). Finally, the leader chooses address (1,1) for the third device in its vector VARL_Leader . This node is denoted as node (1, 1). Generally, the rules for address deriving and assigning can be defined as follows: the leaser chooses for itself address (i, j), for the first two nodes in its vector VARL_Leader (i+l, j) and (i, j+1), and for the third node in its vector VARL_Leader (i+l, j+1) (step S124 in Fig. 9). Thus, the addresses for the nodes from the leader's vector VARL_Leader are derived from the leader's own address by incrementing the x- and/or y-coordinates.

After address assignment to the first nodes in the leader's vector

VARLLeader, a next node is selected for further prosecution (step S126 in Fig. 9): as shown in Fig. 4, different selection schemes for the next node may be applied. For example, one direction may be selected as dominant, and the node in this direction may be selected as the next node. In Fig. 4, the left scheme selects as dominant direction the y-direction, while in the right scheme the x-direction is the dominant direction. In the middle of Fig. 4, the diagonal direction is the dominant direction.

After selection of the next device, the algorithm continues with the following steps s = 1, 2, ... max (mx, my):

4) For step s with 1 < s < max(mx , my ), the algorithm works in an iterative manner as follows:

If s < mx then node(s, 0) configures the two devices in its vector VARLJ that have not received an address so far (if any). The device with highest measured intensity receives address node(s+l, 0) and the other one address node(s + 1, 1). Node (s, 0) addresses them based on their MAC addresses.

If s < my then node (0, s) configures the two devices in its vector VARLJ that have not received an address so far (if any). The device with highest intensity receives address node (0, s+1) and the other one address node(l, s + 1). Node (s, 0) addresses them based on their MAC ad- dresses.

for(i = 1; i < s; i ++) - If s < mx then node(s, i) configures the device in its vector VARLJ that has not received an address so far (if any). This device receives address node (s + 1, i+l) and node node (s, i) communicates with it based on its MAC address.

- If s < my then node (i, s) configures the device in its vector VARLJ that has not received an address so far (if any). This device receives address node (i + 1, s+1) and node node (i, s) communicates with it based on its MAC address.

An algorithmic description could be as follows:

Step s {

node (s, 0) ->

config( s+1. 0)

config(s+l, 1)

node (0, s)

config(0, s+1)

config(l, s+1)

for (i=l, i<s, i++){

node(s, i) ->

config(s+l, j+1)

node(i,s) ->

config(i+l,s+l)

} }

If all devices have assigned addresses, the algorithm stops (step S128 in Fig. 9). Fig. 5 shows an example of the steps of the address assignment to nodes in the grid.

The algorithm described above can run in a distributed or centralized fashion at different moments. The first two phases of the algorithm can be triggered by the commissioning tool, thus, in a centralized manner. The remaining phases (and steps) can operate in a distributed manner. Some of these phases might also be implemented in a centralized manner. For instance, the commissioning device might gather the amount of light that each device j sensed from another luminary jO. Once the commissioning device has collected this information, the commissioning device can locally run the algorithms that allow determining the relative positions of each node in the network.

In the previous section, the distribution of addresses to luminaries arranged according to a square grid according to an embodiment of the inventive algorithm was described. For many settings hexagonal grids offer better features. For instance, LEDs can be packaged better. Another reason is that for the same energy consumption and the same average light intensity, the standard deviation of the light generated by means of luminaries distributed according to a hexagonal grid is around 30 % lower.

In the following, an embodiment of the inventive algorithm for the automatic distribution of addresses in hexagonal grids is described. Fig. 6 depicts such a hexagonal grid, i.e. a grid with hexagonally arranged nodes. Basically, any node can be chosen as the one starting the algorithm (step S120 in Fig. 10). That node picks up address (0, 0) for itself (step S122 in Fig. 10). This node is called node A. Node A uses its vector to look for its six closest neighbors and takes one of them (step S123 in Fig. 10). This node is denoted Node B and receives address (1, 0) (step S123 in Fig. 10). Now node A and B look for its two common direct neighbors and randomly choose one of them that receives address (0, 1) (step S127 in Fig. 10). This node is denoted node C. Then, a further node is selected for address assignment (step S129 in Fig. 10) and it is checked whether each node has an assigned address (step S130 in Fig. 10). If each node in the hexagonal grid has an assigned address, the algorithm stops.

The selection of a further node can be performed as follows: if node A, B, and

C have assigned addresses, node A and node B; B and C; and C and A can configure the node that is opposite node C, A, and B, respectively. In general, the following rule can be used to assign addresses : If node i and node j are direct neighbors with addresses (x_i , y_j ) and (x_j , y_j ), and both nodes i and j have two common neighbors v and u, and node v is assigned address (x_v , y_v ), then node u receives address (x_u , y_u ) where there are three different cases:

{ Case 1 }

if (x_i ! = xj ) & &(y_i == yj

{ Case 1 }if (x_i < x_v ) then

Y_u = y_(i - 1)

X_u = max(x_i , xj )

else

y_(i + D

x u min(x_i , x _j )

end if

{ Case 2 }

if (x_i == xj ) & &(y_i ! = y J ) then

if (x_i < x_v ) then x_u = x_(i - 1)

y_u = min(y_i , yj )

else

x_u = x_(i + 1)

y_u = max(y_i , yj )

end if

{ Case 3 }

if (x_i ! = xj ) & &(y_i ! = y_j ) then

if (x_i = x_v ) then

x_u = x_(i + 1)

y_u = y_i

else

x_u = x_(i - 1)

y_u = y_i

end if

In a further embodiment, luminaries can also be distributed according to a nested grid. For such a case, an algorithm is in the following described that allows allocating addresses to the devices in an automatic fashion.

First, a nested grid is defined as a grid in which a number of luminaries are organized according to a number of L levels. For instance, at level_l = 1 a square grid can be found containing M lighting devices. At level_l = 2, M square grids each around a luminary assigned to level_l can be found, and so on. Such a nested configuration is especially relevant for office lighting in which different types of luminaries are arranged in different positions. The address assigned to a node can comprise a number of identifiers. In general, a device at level_l can have an address including 1 identifiers in which each of the identifiers corresponds to its address at level_l in the nested grid. For instance, if a nested square grid comprising two levels is considered, the luminaries at level_l will receive an address in the form (x_l , y_l ) while another device at level_2 receives (x_l , y_l ); (x_2 , y_2 ).

In this setting, the algorithms described above can be easily adapted to address hierarchical distributions. In general, each luminary is classified according to its type. A luminary type can be further linked to an associated hierarchical level. Given this

information, the algorithm for address allocation runs in L rounds where L corresponds to the number of levels. During round_l = 1 the luminaries at level_l = 1 make use of one of the algorithms above to distribute addresses. Once the first round is finished, each of the luminaries at level_l = 1 looks for devices associated to level_l = 2 in close proximity. Then it starts the address assignment process for them. This process can be repeated until all the luminaries have received an address.

The invention can be applied to networked control systems, particularly networked lighting systems, and used to improve the commissioning and address assignment of devices of the system. Particularly, the invention can help to reduce the commissioning effort, remove errors, and provide more advanced lighting functionalities in combination with an efficient control mechanism. The inventive methods, protocols and algorithms can be the basis for a (set of) standardized command(s) in a lighting control standard such as ZigBee™ or 6L0WPAN. Many deployment scenarios would benefit by the availability of such methods, protocols and algorithms.

This invention has focused on the distribution of network addresses. The invention can be further applied for the distribution of other networking parameters such as, e.g., routing tables, in a similar way. In particular, the node starting the commissioning process can represent a controller or border router and each iteration in the protocol can update the routing tables to get to that controller. It is worth noting that addresses distributed by this invention are well-arranged allowing for a very simple implementation of routing protocols according to , e.g., the address in a grid.

At least some of the functionality of the invention may be performed by hard- or software. In case of an implementation in software, a single or multiple standard microprocessors or microcontrollers may be used to process a single or multiple algorithms implementing the invention.

It should be noted that the word "comprise" does not exclude other elements or steps, and that the word "a" or "an" does not exclude a plurality. Furthermore, any reference signs in the claims shall not be construed as limiting the scope of the invention.

Claims

1. A method for automatically commissioning of devices of a networked control system, which comprises several devices, wherein each device is able to communicate with other devices via a communication means, wherein the commissioning comprises the following steps:

- initiating a first phase for gaining local information in the system, wherein each device detects and identifies its closest neighbor devices and stores the closest neighbor devices (S10), and thereafter

- initiating a second phase for gaining global information in the system, wherein one after another device assigns one or more configuration parameters, which are related to positions in the network, to one or more of its stored closest neighbor devices that have not yet assigned configuration parameters (S12).

2. The method of claim 1, wherein the first phase (S10) comprises the acts of

- transmitting a reference signal by the first device with its communication means (S100), wherein the reference signal contains a device identifier of the first device,

- measuring a parameter of the reference signal received by second devices being neighbored to the first device (S102),

- transmitting the measurement with a device identfier by each second device to the first device using the device identfier of the first device (S104),

- sorting the received device identfiers of the second devices by the measurements (S106), and

- storing the sorted device identifiers of the second devices in the first device (S108).

3. The method of claim 2, wherein the communication means comprise - a light source and the transmitting of the reference signal comprises broadcasting a coded light signal containing the device identifier of the first device,

- RF means and the transmitting of the reference signal comprises broadcasting a RF signal containing the device identfiier of the first device, and/or - ultrasound means and the transmitting of the reference signal comprises broadcasting an ultrasound signal containing the device identfiier of the first device.

4. The method of any of the preceding claims, wherein the devices are arranged in a square grid and the second phase (S12) comprises the following steps:

a) selecting a device in the grid (S120),

b) assigning as configuration parameter an address, which is related to an initial position in the grid, to the selected device (S122),

c) assigning as configuration parameters addresses derived from the address of the selected device to at least one of its stored closest neighbor devices that have not yet an assigned address (S124),

d) selecting a next device from the stored closest neighbor devices of the selected device (S126),

e) repeating steps c) and d) until each of the stored closed neighbor device of the selected device has an assigned address (S128).

5. The method of claim 4, wherein in step d) a next device with the highest measured parameter from the stored closest neighbor devices of the selected device is selected.

6. The method of claim 4 or 5, wherein in step c) addresses derived from the address of the selected device are assigned to the first two devices of the stored closest neighbor devices if these device have not yet an assigned address.

7. The method of any of the preceding claims, wherein the devices are arranged in a hexagonal grid and the second phase comprises the following steps:

a) selecting a device in the grid (S120),

c) selecting a next device from the stored closest neighbor devices of the selected device and that does not yet have an assigned address (S123),

d) assigning as configuration parameter an address derived from the address of the selected device to the device selected as next device (S125), e) assigning as configuration parameter an address derived from the addresses of both lastly selected devices to a further device that is stored as closest neighbor device in both lastly selected devices and has not yet an assigned address (S127),

f) selecting the further device (S129), and

g) repeating steps e) and f) until each of the devices of the grid has an assigned address (S130).

8. A computer program enabling a processor to carry out a method according to any of the preceding claims.

9. A record carrier storing a computer program according to claim 8.

10. A computer programmed to perform a method according to any of claims 1 to 7.

11. A commissioning tool (14) for a networked control system comprising

- processing means being configured to communicate a request to devices of the networked control system to perform a method of any of claims 1 to 7.

12. A device (22) of a networked control system (10) comprising

- communication means (24) for communicating with other devices of the networked control system,

- measuring means (24) for measuring a parameter of reference signals received from other devices of the networked control system,

- storage means (28), and

- processing means (26) being configured

- to sort device identfiers received from other devices of the networked control system together with measured parameters of a broadcasted reference signal by the measured parameters and to store the sorted device identifiers together with measured parameters in the storage means,

- to perform a method of any of claims 2 to 7.

13. A system for automatically commissioning of devices of a networked control system comprising

- several devices (22) according to claim 12, and

- a commissioning tool according to claim 11 for initiating a first phase for gaining local information in the system, wherein each device detects and identifies its closest neighbor devices and stores the closest neighbor devices (S10), and thereafter initiating a second phase for gaining global information in the system, wherein one after another device assigns one or more configuration parameters, which are related to positions in the network, to one or more of its stored closest neighbor devices that have not yet assigned configuration parameters (S12).