FIELD OF THE INVENTION

The present invention generally relates to the field of networked lighting systems and methods, devices and luminairies corresponding thereto.

BACKGROUND OF THE INVENTION

Networked lighting systems comprise luminaires, lighting sensors, and motion sensors, interconnected by means of some communication medium. The communication medium may be based on wires, radio frequency (RF) technology, or visible light communications (also known as coded light).

The interconnection between the luminaires, lighting sensors, and motion sensors enables a controlled lighting system, in which the information sensed by the sensors may be used to control the luminaires and, in particular, determine the intensity settings of these luminaires.

Known functionalities offered by these networked lighting systems are presence control, whereby luminaires, based on data sensed by motion sensors, are switched on only when people are present, and daylight adaptation, whereby luminaires, based on data sensed by light sensors, are dimmed when daylight enters the room (e.g. through the windows of the room). Thereby, luminaires need not be switched to full power to provide the users with the desired illumination levels. Alternatively, such systems may offer the control functionality to create some arbitrary, useful, light distributions.

Typically, one sensor controls one or more luminaires. Moreover, each of the luminaires controlled by the one sensor is set to the same targeted intensity. Traditionally, the settings of such a group of luminaires are controlled by means of thresholds. If the light sensed by the light sensor falls below a low threshold, all luminaires controlled by this light sensor are set to a high target value. Conversely, if the illuminance sensed by the light sensor rises above a high threshold, the lights are dimmed to a low value. If the intensity sensed by the sensor is between the two intensity thresholds, the light setting of the luminaires controlled by this light sensor is kept unchanged.

SUMMARY OF THE INVENTION The present invention generally concerns daylight adaptation and light control in a networked lighting system based on illuminance sensing. It is an object of the present invention to overcome these problems, and to provide improved control of the light setting of the luminaires in a networked lighting system. To achieve improved control of the light setting of the luminaires in a networked lighting system the disclosed system therefore preferably deploys at least a number of illuminance sensors and luminaires interconnected by some interconnection technology, in addition to other devices that may be provided in the system.

According to a first aspect of the invention, the above and other objectives are achieved by a method for setting light intensity settings of light sources in a networked lighting system, comprising determining a mapping∑ from intensity settings Js as obtained by N light sensors in the networked lighting system to intensity settings I _I for M light sources in the networked lighting system; combining the mapping∑ with desired lighting settings T as desired by the N light sensors, thereby forming an intensity vector∑T; and transforming the intensity vector∑T to dimming levels I _∑ for the M light sources.

Control overhead for adjusting the light source one at the time is therefore removed or at least reduced. Advantageously this smoothes the light intensity values.

Thereby blocking artefacts may be avoided. This provides a lighting effect that is more pleasing to the human eye.

The method may further comprise setting light intensity levels of the M light sources in the networked lighting system according to the dimming levels I _∑ . Adjustments made to the light sources in the networked lighting system may thus be based on the determined light intensity levels. Manually fine tuning each light source would thereby no longer be required.

The desired lighting settings T may be determined based on currently sensed intensity values by the N light sensors. Further, the desired lighting settings T may be based on comparing the currently sensed intensity values by the N light sensors to one or more threshold values, preferably two or more threshold values.

The mapping∑ may be based on a transfer matrix D, a location matrix X, a sound matrix S, and/or a received signal strength matrix R. Allowing the mapping∑ to be determined based on one or more properties may advantageously enable the mapping∑ to be optimized for a given context and for a given set of parameters (such as a transfer function, location properties, sound properties and/or signal strength properties). The dimming levels I _∑ at time n+1 may further be determined as a weighting between∑T and the dimming levels I _∑ at time n. Thereby the dimming levels may be smoothed over time.

The method may further comprise communicating light intensity settings between the M light sources. This may allow for a distributed implementation of the control.

According to a second aspect of the invention, the objective is achieved by a computer program product comprising software instructions that when downloaded to a computer is configured to perform a method according to the first aspect.

According to a third aspect of the invention, the objective is achieved by a non- volatile storage medium comprising a computer program product according to the second aspect.

According to a fourth aspect of the invention, the objective is achieved by a control device for setting light intensity settings of a light source in a networked lighting system, comprising a receiver arranged to receive light intensity settings Js as obtained by N light sensors in the networked lighting system; a processing unit arranged to determine a subset∑' of a mapping∑ from the received intensity settings to intensity settings I _I for M light sources in the networked lighting system; the processing unit further being arranged to combine the subset∑' with a subset T' of desired lighting settings T as desired by the N light sensors, thereby forming an intensity parameter Σ'Τ'; and the processing unit further being arranged to transform the intensity parameter Σ'Τ' to a dimming level I _∑ - for one of the M light sources. The control device may further be arranged for setting light intensity settings of M light sources in the networked lighting system, wherein the subset∑' corresponds to the entire mapping∑ and , wherein the subset T' corresponds to the entire desired lighting settings T, whereby the entire mapping∑ and the entire desired lighting settings T thereby form an intensity vector∑T; and wherein the processing unit further is arranged to transform the intensity vector∑T to dimming levels I _∑ for the M light sources in the networked lighting system.

According to a fifth aspect of the invention, the objective is achieved by a luminaire comprising at least one light source and a control device according to the fourth aspect, the control device being arranged for setting light intensity settings at least of the at least one light source.

It is noted that the invention relates to all possible combinations of features recited in the claims. Likewise, the advantages of the first aspect applies to the second, third, fourth and fifth aspects, and vice versa. BRIEF DESCRIPTION OF THE DRAWINGS

The above and other aspects of the present invention will now be described in more detail, with reference to the appended drawings showing embodiment(s) of the invention.

Fig. 1 illustrates a control system according to embodiments;

Fig. 2 illustrates a luminaire according to embodiments;

Fig. 3 illustrates a controller according to embodiments; and

Fig. 4 is a flowchart of a method according to embodiments.

DETAILED DESCRIPTION

The below embodiments are provided by way of example so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. Like numbers refer to like elements throughout. The devices disclosed in the below embodiments will be described in an operation context.

According to the state of the art, typically one or more luminaires is controlled by means of one sensor. Moreover, each of the luminaires controlled by the one sensor is set to the same targeted intensity.

A control system 1 is illustrated in Fig. 1. The control system 1 of Fig. 1 represents an example layout of a networked lighting system with eight light sources 2a-h and two sensors 3a-b. Assume as a first example that a currently deployed control algorithm (for example as executed by the control device 4) provides the following dimming values for the light sources 2a-h:

Thus, with reference to the example layout of the control system 1 , the light sources 2a-d close to sensor 3a are all set to a high level (i.e. 100%), and the light sources 2e- h close to sensor 3b are all dimmed to a low level (i.e. 50%>). This gives a sharp contrast, which may not be pleasing to the eye of a person observing the collected light output of the light sources 2a-h. Thus, setting the light intensity values according to this process generally results in patchy light settings. According to the present example, groups of light sources (light sources 2a-d and light sources 2e-h, respectively) can be distinguished that have the same intensity values, with very abrupt changes in the light settings of the light sources in the various groups. Generally, this is displeasing to the eye. It is also reminiscent of the blocking artefacts distorting digital images in digitally coded image streams. With the advent of LED lighting, lighting control systems 1 are likely to comprise many more luminaires, other light sources and lighting devices. Therefore, lighting control systems will become more screenlike, and the artefacts will be even more perceptible.

It is possible to change the light intensity settings of the individual light sources by modifying the high and low intensity values that are used in the control command to some smoother distribution. However, this distribution must be commissioned.

Additionally, this involves identifying exactly which light source is to be switched to which intensity. This process can be both cumbersome, time consuming, and error prone.

Additionally, the settings preferably must accord well with the intensity settings of the light sources controlled by other sensors, and must preferably therefore be made dependent of the actual luminance values sensed by these sensors. This again requires non-negligible control overhead, in addition to the commissioning and design effort.

The networked lighting system of the present invention provides settings of light intensity values whereby blocking artefacts are avoided, thus resulting in smooth transitions between light intensity values of neighboring light sources. According to the networked lighting system of the present invention the control loop is adapted so as to provide smooth light source intensity settings, thus avoiding the above mentioned drawbacks of the prior art process.

Fig. 2 schematically illustrates a light source 2a, 2b, 2c, 2d, 2e, 2f, 2g, 2h. The light source 2a-h comprises a receiver 6, a light driver 7 and a light emitter 5. Optionally the light source 2a-h may further comprise a processing unit 8 and a memory 9. The receiver 6 is arranged to receiver measurements and/or control signals. The light driver 7 is arranged to provide the light emitter 5 with a driver signal corresponding to a dimming level. The processing unit 8 may be arranged to perform calculations and determine light intensity settings for the light source. The memory 9 may hold instructions relating to instructions performed by the processing unit 8.

Fig. 3 schematically illustrates a control device 4. The control device 4 comprises a receiver 10 and a processing unit 12. Optionally the control device 4 may further comprise a memory 11. The receiver 10 is arranged to receiver measurements and/or control signals. The processing unit 12 may be arranged to perform calculations and determine light intensity settings for one or more light sources 2a-h. The memory 11 may hold instructions relating to instructions performed by the processing unit 12. Preferably the memory 11 is non-volatile. Assume a networked lighting system 1 comprising N number of sensors and M number of light sources. Starting generally, first an M by N mapping∑ is defined. In a step S02 the mapping∑ is thus determined. The receiver 6 of the light source 2a-h may be arranged to receive light intensity settings. The processing unit 12 may be arranged to determine at least a subset∑' of the mapping∑ based on the received light intensity settings. In a centralized implementation the subset∑' preferably corresponds to the entire mapping∑ and is preferably determined by the processing unit 12 of the control device 4. The mapping ∑ may be regarded as a smoothing matrix and maps from intensity settings Js as obtained by each of the N sensors 3a-b to intensity settings I _I for each of the M light sources 2a-h. In other words∑ may be described as a mapping:

II =∑ Js

There may be different ways of determining the mapping∑, for example by determining a transfer function, a location function, a sound function, signal strength or a combination of two or more of these parameters.

According to one embodiment the mapping∑ is based on a transfer matrix D in which entry D(s,l) describes intensity effect of light source 1 on light sensor s. In more detail, the transfer matrix D measures the influence of a light source 1 on each of the N sensors. The intensity settings of the light sources 2a-h may thus be smoothed based on this system matrix D, given the intensity settings per sensor 3a-b. Specifically, the intensity of the light sources 2a-h may be set as

II =∑ Js = Δ ¹ D ^l Js

where Δ is a, normalizing, diagonal matrix, which has as entries the column sums of D:

Δ = D ^l 1,

where 1 is an N-length vector of ones, and the mapping thus is∑ = Δ ^"1 D ^l . D may be measured in an automatic darkroom calibration procedure. In this procedure, light sources 2a- h are successively switched on, and the corresponding illuminance is measured at all sensors

3a-b in the networked lighting system 1.

From a physical point of view this means that a matrix is determined in which each row contains the relative effect of each light source 2a-h on each of the N sensors 3a-b in the networked lighting system 1. Thus, the smoothing operation is according to this embodiment based on the transfer matrix D, and the mapping∑ may be directly determined from said transfer matrix D. As the transfer matrix D can be measured by means of an automatic procedure, this smoothing procedure can be made fully automatic. Additionally, as the transfer matrix D preferably must be known anyway to properly design the networked lighting system (for example to avoid oscillations), there is generally no additional overhead in deriving D.

According to one embodiment the mapping∑ additionally or alternatively is based on a location matrix X in which entry X(s,l) describes a distance between light source 1 and light sensor s. The locations of the light sources 2a-h and the sensors 3a-b may thus be used to define a distance between each sensor 3a-b and each light source 2a-h. The distance values are then combined to form the location matrix X. Another approach is to define a smooth parametric model in terms of the spatial coordinates of the light sources 2a-h. The light sources 2a-h may then be dimmed to the intensities that are closest to the intensities determined for T and I (see below) subject to the constraints imposed by the parametric model.

According to one embodiment the mapping∑ additionally or alternatively is based on a sound matrix S in which entry S(s,l) describes a sound effect of light source 1 on sound sensor s. For example, as an extension of using a transfer matrix D or a location matrix X, other methods can be used to determine information relating to the distance between the light sources, e.g., based on sound.

According to one embodiment the mapping∑ additionally or alternatively is based on signal strength, when the desired settings T are transmitted via radio frequency (RF) signals between the light sources 2a-h and/or sensors 3a-b. The mapping∑ may thus be based on a received signal strength matrix R in which entry R(s,l) describes a received signal strength of a RF transmitter at light source 1 on a RF receiver at sensor s.

Once the mapping∑ has been determined, it is combined with desired lighting settings T as desired by the N light sensors, step S04. The processing unit 8 of the light source 2a-h may be arranged to combine at least the subset∑' with a subset T' of the desired lighting settings T, thereby forming an intensity parameter Σ'Τ'. In a centralized

implementation, preferably implemented in the processing unit 12 of the control device 4, the subset T' preferably corresponds to the entire lighting settings T and thus the combination∑T forms an intensity vector. The desired lighting settings T are preferably determined based on currently sensed intensity values by the N light sensors. Denote by T ⁽ⁿ⁾ the desired lighting settings T at time n and by T^ ¹ -* the desired lighting settings T at time n+1. For example, the desired lighting settings may change from time n to time n+1 according to the following expression:

_T (n+D _{= R} . _1(E < _0i ) _{+ L} . _1(E > _0h ) _{+ T} (n) . j^ < £ < 9 _h ) In this expression H is a first setting value. L is a second setting value lower than the first setting value H. E represents currently sensed intensity values by the N light sensors, θι is a first threshold value. 9 _h is a second threshold value higher than the first threshold value θι. Here, the indicator function 1 (i.e. the index next to H, L, and in the expression for T ⁽ⁿ⁺¹⁾ ) maps a vector of N Booleans to a N-length vector of zeroes and ones. In other words, H and L are thus the high and low intensity values, respectively; the light source intensities are set to the high value, H, when the light sensed is below the low threshold θι, the light source intensities are set to the low value, L, when the light sensed exceeds the high threshold 9 _h , and the light source intensities are left unchanged if the light intensity sensed by the sensor falls between these thresholds. As the skilled person understands, the expression for T ⁽ⁿ⁺¹⁾ is extendable to multiple intermediate levels by the use of further thresholds.

According to the above expression for T ⁽ⁿ⁺¹⁾ the intensity settings for each sensor 3a-b are thus determined. Multiplying the vector T ⁽ⁿ⁺¹⁾ with the mapping∑ then derives the vector of intensity settings of the individual light sources 2a-h.

The combination of∑ with T thus forms an intensity vector∑T. The intensity vector∑T is then, in a step S06, transformed to dimming levels I _∑ for the M light sources in the networked lighting system 1. The processing unit 8, 12 of the light source 2a-h or the control device 4 may thus further be arranged to transfer the intensity parameter Σ'Τ' (or the intensity vector∑T) to a dimming level I _∑ > (or dimming levels I _∑ ) for one (or all) of the M light sources 2a-h in the networked lighting system 1. For example, the intensity vector∑T may be projected onto the set of possible values by a mapping P:

I _∑ = P (∑T)

The mapping P may thus ensure that the intensity vector∑T is mapped to the range [0, 100%] which thus may be used as possible dimming levels by the M light sources 2a-h. The light intensity levels of the M light sources 2a-h in the networked lighting system 1 may then, in a step S08, be set according to the dimming levels I _∑ . The light driver 7 of the light source 2a-h thus transmits a driver signal corresponding to the dimming level to the light emitter 5 of the light source 2a-h, thereby setting the dimming level of the light source.

As noted above, the disclosed smoothing process may either be centralized or distributed. For example, in the centralized approach the central control device 4 may have access to all matrices and/or measurements. Further, in the centralized approach the central control device 4 preferably performs the calculations to determine∑, T and I _∑ . Further, in the centralized approach the central control device 4 preferably controls the light sources 2a-h to dim to the determined intensities. For a distributed implementation, the determinations for∑, T and/or I _∑ are decoupled. The first operation of the distributed implementation determines, by the processing unit 8 of every light source 2a-h, the target settings as in the above expression for j(n+i) f _{or eac} ] _{1 sensor} 3a_b it is natural that each light source 2a-h adapts the target settings for a given sensor each time new sensor information from this sensor 3a-b becomes available. Based on these target settings for the sensors 3a-b, the light source 2a-h derives a target setting for itself, based on the above expression for I _∑ . For this, a light source 1 must have access to row 1 of the mapping∑.

To gain the information in a row 1, light source 1 preferably has to shine at a reference light intensity during a calibration phase. Each sensor s then preferably senses the received intensity I _ljS and sends it to light source 1. The row 1 then comprises all the M received values normalized by the sum over all s of the values I _ljS . In order to achieve this, the light sources 2a-h may be provided with a light intensity sensor. Furthermore, the light sources 2a-h may be arranged to communicate (e.g. via visible light communications, or so- called coded or modulated light) light intensity settings, step S10. The light intensity sensors can be used both to sense the light intensity, and to decode the coded light messages.

Each light source 2a-h may be arranged to embed the desired intensity setting T in a broadcast message which is transmitted via coded light. This broadcast preferably also includes a (unique) identity of the transmitting light source, and its intensity I.

Each light source 2a-h may then determine a smoothed intensity setting I by taking a weighted sum of the received desired intensity settings in these broadcast messages, where the weights may be defined on the basis of the received signal strength of the broadcast divided by the intensity I included in the broadcast.

The ratio between the received signal strength and the intensity I included in the broadcast represents a given entry of the mapping∑. Since this value might oscillate due to external factors such as reflections or noise, a light source 2a-h may be enabled to determine a better estimation of this value by averaging over the last K messages. The average can also be weighted by means of a given window to give more or less importance to the last samples. A further type of smoothing thus refers to smoothed control over time.

Smoothed control over time is implementable in a central control device 4 as well as in a distributed implementation. It may for example be preferred that the light intensity settings do not change too rapidly over time since fast changes may disturb the users. It may thus be advantageous to allow also for temporal adaptation in relation to a change in the light emitted by the light sources 2a-h. The temporal adaptation may, for example, be implemented by means of a memory factor a, where 0 < a < 1. The intensity setting may be given as

I _∑ ⁽ⁿ⁺¹⁾ = a P (∑ T ⁽ⁿ⁺¹⁾ ) + (l- ct) I∑ ⁽ⁿ⁾

where Ι _Σ ^ ¹ ^ is the determined intensity at time n+1 and is the determined intensity at time n. Thus the memory factor a may act as a weighting factor between the projected target setting P (∑ T^ ¹ -*) and the currently used intensity settings l _∑ - ⁿ Thereby the dimming levels I _∑ at time n+1 may further be determined as a weighting between∑T and the dimming levels I _∑ at time n. The memory factor a is preferably chosen so that the intensities change in a perceptually pleasant way, or, e.g. such that the adaptation is not perceivable by a human observer.

A configuration of N=2 sensors and M=8 light sources as in Fig. 1 will first be considered. The results will be compared to those of the first example as disclosed above. As in the first example it will also in the present second example be assumed that there are two non-zero dimming levels for each light source, namely 100% (high intensity) or 50% (low intensity).

The transfer matrix D for this configuration takes the following values:

Light sources 2a through 2d are associated with sensor 3a, and light sources 2e through 2h are associated with sensor 3b. According to the present second example the system 1 is so configured that intensity setting are desired, depending on the sensor input, of 100%) for light sources 2a through 2d and 50%> for light sources 2e through 2h. Applying the transfer matrix D as outlined above yields the following dimming values for the light sources 2a-h:

In comparison to the dimming values for the light sources 2a-h according to the first example, the light settings display a gradient and that the blocking artefact has been considerably reduced.

If instead a location matrix X is used to determine∑ the following dimming values for the light sources 2a-h are obtained: Dimming 100% 100% 85% 85% 65% 65% 45% 45%

As a third example, the same light source configuration as in the first and second examples is considered. However, in the present third example it is further assumed that each light source comprises a sensor. Hence N=M=8. Further it is assumed that the following dimming values are desired for the light sources 2a-h:

Applying a location matrix X to determine∑ for the present third example results in the following dimming values for the light sources 2a-h:

Hence, the second and third examples shows that sharp contrasts in dimming levels are avoided by smoothing the dimming levels.

The person skilled in the art realizes that the present invention by no means is limited to the preferred embodiments described above. On the contrary, many modifications and variations are possible within the scope of the appended claims. For example, the present invention may advantageously be applied to lighting products for both indoor and outdoor environments. For indoor environments, applications are for example in the fields of intelligent offices. For outdoor environments, the proposed embodiments may

advantageously allow for efficient and smoothed light on demand. Furthermore, the implementations of the present invention can be based on a number of existing standards and technologies such as coded light; 6L0WPAN/C0AP, or ZigBee. The disclosed method may preferably be provided as a set of software instructions thus forming a computer program product. When downloaded to a computer the computer program product is thus configured to perform the disclosed method. Preferably the computer program product is stored on a non- volatile storage medium 9, 11. A luminaire may comprising at least one light source 2a-h as disclosed above. The luminaire may further comprise the disclosed control device 4.