Directional climate control system with infrared target search

Abstract

Method for controlling a heating, ventilation and air conditioning system, HVAC system, (20) with a steerable outlet (13-17) for introducing a stream of treated air into the passenger compartment (10) of a vehicle, comprising the steps: Acquisition of thermographic images (75) covering a fixed region in the passenger compartment (10) in which an occupant (25-28) is potentially located; Compressing a thermographic image (75) to a temperature map (76) representing pixels (123-127) of the thermographic image (75) which fall within a specified temperature range corresponding to the occupant (25-28); Filtering the temperature map (76) according to a sliding window (122) to combine continuous regions (131) of pixels (123-127) that on average fall within the specified temperature range; Quantifying an area for each continuous region (131); Locating a focal point (132) of a continuous region (131) with a largest area; and Directing the steerable outlet (13-17) towards the center of gravity (132).

Description

BACKGROUND OF THE INVENTION

[0001] The present invention relates generally to automotive HVAC systems and, more particularly, to the automatic adjustment of steerable air outlets to direct treated air to optimize passenger comfort and reduce energy consumption.

[0002] Heating, ventilation, and air conditioning (HVAC) systems control the climate in transportation vehicles such as automobiles to maintain the thermal comfort of the vehicle occupants. Typically, a fan forces air through heat exchangers, delivering conditioned air to various points within the passenger compartment. For example, warm air may be provided by a heater core, drawing heat from coolant flowing within an internal combustion engine. Cool air may be drawn using a compressor mechanically driven by the engine. Electrified vehicles such as hybrid electric vehicles (HEVs) and battery electric vehicles (BEVs) may instead use electrically powered devices such as PTC resistance heating elements and electric air conditioning compressors.

[0003] The simplest automotive climate control systems allow the occupant to directly control the intensity of heating or cooling, the operating speed of the fan, and the relative amount of air delivered to various vents. This requires the user to constantly monitor and adjust the climate control settings to stay within a comfortable range. However, vehicle occupants may not know how best to control the HVAC system to optimize its efficiency. They may turn the control knobs to maximum power while simultaneously directing the fan blades up or down, away from their bodies, forcing conditioned air onto less critical surfaces and requiring more energy to provide occupant comfort. This also results in the fan being louder than necessary to achieve the desired thermal comfort.

[0004] Electronic Automatic Temperature Control (EATC) systems have also been introduced, in which a control system monitors the ambient air temperature in the passenger compartment and automatically adjusts the fan speed and heater core or air conditioning operation to maintain the desired temperature setting. Some vehicles implement multiple zones of separate automatic temperature control, with individual target temperature settings for each zone.

[0005] Traditional HVAC systems only indirectly control an occupant's actual skin temperature. Because skin temperature is a better indicator of actual occupant comfort, systems for controlling HVAC system operation have been investigated that use infrared (IR) sensors to determine the skin temperature of vehicle occupants and then adjust a HVAC system temperature setpoint as needed to achieve a target skin temperature. However, the thermodynamic environment inside a vehicle is complex, as are the relationships between different HVAC control settings and the resulting effect on the skin temperature of different occupants. Therefore, previous systems have been relatively complex and not cost-effective.

[0006] To reduce energy consumption and ensure optimal comfort, various systems have been proposed that automatically adjust the pattern of airflow delivered to the passenger compartment based on seat occupancy (e.g., by turning off vents when a seat is unoccupied). However, to date, no known system has provided sufficient accuracy or performance to truly optimize the directional adjustment of treated air over a sufficiently wide range of thermal conditions.

[0007] JP 2001-349 786 A relates to a calibration method for a non-contact temperature sensor for vehicles. In one embodiment, the output signal of the non-contact temperature sensor is processed in a signal processing circuit to determine temperature distribution data. This data is then transmitted via the output signal processing to a control circuit, which operates the vehicle equipment (air conditioning system, airbag device, safety control device, etc.) based on the temperature distribution data from the signal processing circuit. DE 10 2013 204 988 A1 relates to the personalized control of an HVAC system based on the skin temperature of a vehicle occupant or driver. DE 10 2005 017 829 A1 relates to an air conditioning system for a motor vehicle in which a passenger compartment is air-conditioned using a non-contact temperature sensor.From DE 10 2013 001 679 A1 a method and a device for image recording and image evaluation for controlling and / or regulating an air conditioning system are known. SUMMARY OF THE INVENTION

[0008] The present invention utilizes thermographic imaging to identify the skin surfaces of vehicle occupants, which can be targeted with a treated airflow using electronically controlled vents featuring dual-axis directional control and optional nozzle control for concentrating the airflow. Based on the detected seat occupancy, airflow from multiple vents can be directed to detected skin targets, depending on which seats are occupied. Thermographic images can be captured sequentially to update the airflow direction and flow rate under changing conditions.

[0009] According to one aspect of the invention, a method for controlling a heating, ventilation, and air conditioning (HVAC) system having a steerable outlet for introducing a flow of treated air into the passenger compartment of a vehicle is presented. A corresponding vehicle device for carrying out a method according to the invention therefore comprises a heating, ventilation, and air conditioning (HVAC) system having a steerable outlet for introducing a flow of treated air into the passenger compartment of a vehicle. A thermographic imager is configured to capture thermographic images covering a fixed region in the passenger compartment potentially containing an occupant.The HVAC controller is configured to a) compress a thermographic image into a temperature map representing pixels of the thermographic image falling within a predetermined temperature range corresponding to the occupant, b) filter the temperature map according to a sliding window to merge contiguous regions of pixels that, on average, fall within the predetermined temperature range, c) quantify an area for each contiguous region, d) locate a centroid of a contiguous region having a largest area, and e) direct the steerable outlet toward the centroid. BRIEF DESCRIPTION OF THE DRAWINGS Fig. is a perspective view of a motor vehicle passenger compartment having an HVAC system with a plurality of steerable air outlets or vents for providing climate control to occupants of the vehicle. Fig. is a block diagram showing a preferred embodiment of a vehicle device configured to carry out a method according to the invention. The Fig. are side views of steerable air outlets. Fig. shows a sequence of image data used in a preferred embodiment for directing a steerable air outlet toward an occupant. Fig. is a flowchart of a preferred embodiment of a method for aligning a steerable air outlet. Fig. is a flowchart of a preferred embodiment for determining target areas. The Fig. are representative diagrams showing predefined temperature ranges for identifying an occupant in a thermographic image. Fig. shows a pixel arrangement for a portion of a thermographic image after conversion to a bivalent temperature map. The Fig. illustrate a sliding window at successive positions during filtering of a temperature map to merge regions in the two-value temperature map. Fig. illustrates a pixel arrangement after the regions have been merged. Fig. illustrates a preferred technique for locating a centroid in a merged region. DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

[0010] Fig. shows a passenger compartment 10 of a motor vehicle having various seating locations, such as a front seat 11 and rear seats 12. A vehicle HVAC system includes a plurality of air outlets or vents 13-17 for introducing respective streams of treated air into the passenger compartment 10. One or more of the vents 13-17 includes a steerable outlet such that the direction and / or focus (i.e., spread) of the HVAC system is automatically adjustable.

[0011] Fig. shows an HVAC system 20 with various components interconnected via an electrical system 21, which preferably includes an electrical wiring harness and a multiplexed communication system as commonly used in the industry. An HVAC controller 22 may preferably comprise a typical electronic module having customized circuitry and / or one or more programmable microcontrollers with appropriate software and / or firmware, as commonly used in motor vehicles.

[0012] A thermographic imager, preferably in the form of an infrared camera 23, is arranged at a location in the front of the passenger compartment to capture thermographic images covering a fixed region in and around the front seating areas 25 and 26. The occupant seating area 26 may, for example, be intended for a driver. The HVAC controller 22 may be configured to always assume that the driver is present. On the other hand, the occupant seating area 25 may represent a passenger seat that may potentially have a passenger present, but may also be unoccupied. Another infrared camera 24 may be arranged at a rear location to capture thermographic images, with the infrared camera covering a rear seat area in the passenger compartment that includes the occupant seating areas 27 and 28.

[0013] In the front seating area, fans 30 and 37 may be provided for the right and left sides of the passenger compartment, which are separately controllable at a variable speed determined by the HVAC controller 22. Fan 30 provides a flow of treated air to controlled vents 31, 33, and 35. Vent 31 is mounted on the instrument panel on the left, outer side of the instrument panel near the occupant seat 25 and serves to provide an air flow 32. Vent 33 is mounted on the left inner side of the instrument panel and provides an air flow 34, and vent 35 is a left floor vent for providing an air flow 36. At least one of the vents 31, 33, and 35 is automatically steerable to direct an air flow to a desired destination in the occupant seating area 25.Similarly, a variable-speed fan 37 is coupled to the front, inboard, right-hand side of the instrument panel, which provides airflow 39, and to the front, outboard, right-hand side of the instrument panel, which provides airflow 41, and to the right-hand floor vent 42, which provides airflow 43 to the passenger compartment 26. Any combination of steerable and fixed vents may be employed in the present invention. Furthermore, a particular steerable vent may be configured to alternatively direct airflow to different passenger compartments.

[0014] A blower 44, also controlled by the HVAC controller 22, delivers treated air streams to rear vents 45, 47 and 49, with air streams 46, 48 and 50 being directed to occupant areas 27 and 28 as shown.

[0015] For the purposes described in detail below, the HVAC controller 22 is connected to, or may include, a lookup table 51. Additionally, the HVAC controller 22 is connected to various sensors, including an ambient / outside air temperature sensor 52, an interior temperature sensor 53, and seat occupancy sensors 55. The HVAC controller 22 is connected to a state-of-the-art human machine interface (HMI) 54 mounted in the passenger compartment. The controlled vents include actuators activated by the HVAC controller 22 based on signals from the HMI 54, sensors 52, 53, and 55, and IR cameras 23 and / or 24 to direct treated air to achieve passenger comfort with optimized energy consumption.

[0016] Fig. shows an example of a steerable air diffuser 60 installed in an instrument panel or dashboard 61. An air duct 62, receiving treated air from a blower, delivers the air to a movable outer shroud 63 of the diffuser, which shroud is supported on a cup-shaped base 64. A gimbal 65, which can be rotated along one or more axes by a motor 66 under the control of the HVAC controller 22, is connected to the shroud 63. By manipulating the gimbal 65, an airflow exits an opening in the shroud 63 and travels in a desired direction. Internal dampers or gates, such as damper 67, may be disposed in the shroud 63 and coupled to an actuator motor 68 to shape (i.e., throttle) the flanks of an existing airflow to provide a desired focus of the outlet airflow.

[0017] Fig. shows an alternative embodiment for directing an exhaust airflow with an vent having controllable vanes 70 and 71. Although two vanes are shown, a larger number of vanes are typically used. Each vane 70 and 71 has end pivots 72 rotatably mounted on a vent housing (not shown). The vanes are all connected by a connecting rod 73 to maintain a parallel spacing. The up and down movement of the rod 73 rotates the vanes 70 and 71 together to direct an exhaust airflow along a controllable direction. The rod 73 may be connected to an actuator motor (not shown) driven by the HVAC controller 22.

[0018] Fig. shows an overall process of the present invention for using a thermographic image to identify a target location for directing an exhaust airflow. A thermographic image 75 is acquired according to a fixed field of view using an infrared camera to output a multi-value image with pixels having a color or intensity representing corresponding temperature values. Based on a predetermined temperature range selected to encompass an expected temperature of exposed skin, taking ambient temperature conditions into account, the multi-value thermographic image is compressed to a two-value image 76 such that pixels falling within the predetermined range are colored white, for example, and all other pixels are colored black.Thus, image 76 comprises a temperature map in which regions of white pixels correspond to exposed skin areas of a vehicle occupant. In the example shown, a head region 81 and arm / hand regions 82 and 83 are recognizable in temperature map 76. However, imaging noise causes regions 81 to 83 to contain scattered black pixels, making it difficult for the controller to automatically detect the region.

[0019] Figure 77 shows another processed image in which regions of pixels falling within the specified temperature range have been merged so that regions become essentially continuous, with well-defined edges and no scattering of black pixels within the regions. As described below, the merge can be accomplished by filtering, using a sliding window to sequentially remap each pixel value to match a majority of the respective surrounding pixels.

[0020] Since the number of identified regions may exceed the number of steerable air outlets, the detected regions can be ranked by size so that the largest region can be selected for target search, as shown in Figure 78. In Figure 79, a center of gravity 88 of each selected continuous region is located to provide a precise target for controlling the corresponding air outlet.

[0021] A preferred method of the invention is described in Fig. , wherein the external ambient temperature and the internal passenger compartment temperature are measured in step 90. Preferably, performance of the method may be initiated by unlocking or opening a door to indicate that a vehicle drive cycle is about to begin. In step 91, an initial thermographic image is acquired, e.g., before an occupant is seated inside the passenger compartment. The initial image allows any particularly prominent infrared "hot spots" to be recorded and identified. In step 92, the HVAC system is activated in response to initial conditions as identified by the temperature measurements from step 90 and the first IR image from step 91.Additionally, the initial activation of the HVAC system can be based on historical HVAC settings for a temperature setpoint and fan speed established by the driver during previous trips under similar conditions, or according to standard control values ​​provided by the manufacturer. The time for HVAC activation can be determined in a conventional manner, such as when a vehicle engine is remotely started via a wireless control or when the driver enters the vehicle and starts the engine or other component using an ignition key.

[0022] In response to the ignition being turned on, a thermographic image is captured with the driver in the appropriate position in step 93. The image in step 93 may potentially include additional occupants who will be present during the upcoming drive cycle. Using the captured thermographic image, the target areas are determined in step 94. The direction and / or focus of various HVAC airflow outlets are adjusted in step 95. Optionally, the HVAC system outlet temperature and fan speed or intensity may also be adjusted. After a predetermined period of time Δt, a new thermographic image may be captured in step 96. Because an HVAC system actively influences the temperature environment, when subsequent thermographic images are captured, the desired skin features in the image may be masked by the appearance of the airflow.This effect is particularly noticeable when the fan is operating at higher speeds. To reduce the presence of a temperature signature of the treated air in the image, a check is performed in step 97 to determine if the fan speed is above a threshold. If not, the target areas in the subsequent image are determined in step 94, and then the HVAC outputs are adjusted in step 95. If the fan speed is above the threshold, the subsequent thermographic image can be filtered in step 98 to remove the effects of the treated airflow in the image. The filtering can be done based on the turbulent nature of airflow seen in a typical thermographic image (e.g., image regions with a certain texture can be deleted from the image).Alternatively, the fan speed can be temporarily reduced or the fan can be switched off completely while the thermographic image is being captured.

[0023] Fig. shows a preferred method for determining target areas in greater detail. In step 100, the measured internal and external ambient temperatures are used to select the predetermined temperature range for identifying regions in the image that show an occupant's skin. The external temperature indicates whether the occupant's skin temperature is raised or lowered by external conditions compared to normal body temperature. The internal passenger compartment temperature helps identify the expected temperatures of interior surfaces in the vehicle (e.g., head restraints). The predetermined temperature range is determined empirically to include the expected skin temperatures while excluding as many of the vehicle's internal surfaces as possible. Preferably, the values ​​of the predetermined ranges are stored in a lookup table located in or accessible to the HVAC controller.

[0024] The lookup table can be organized according to index values ​​for the internal and external temperatures. Fig. shows an example of specifying a temperature range corresponding to a particular external temperature and showing the range of change for a varying internal temperature. The specified range has a maximum value defined by a curve 107 and a minimum value defined by a curve 108. For an internal temperature 112 at the lower end of curves 107 and 108, a specified range 110 is defined by the respective values ​​of curves 107 and 108. For other values ​​of internal temperatures (at the same external temperature), a different specified range can be obtained, for example, range 111 at a higher internal temperature, specified at 113. Fig. is representative of curves obtained for high external ambient temperatures (i.e., situations requiring HVAC cooling). In hot external conditions, the vehicle interior is typically considerably heated, making it easier to distinguish skin temperatures. Therefore, the specified range in region 111 may be wider than in region 110. At more moderate temperatures, it may be more difficult to distinguish skin temperatures from body structures, necessitating a narrower range in 110.

[0025] Fig. is representative of lower external temperatures (i.e., a situation requiring heating of the passenger compartment), with curves 114 and 115 defining the maximum and minimum temperatures of the specified temperature range, respectively. As already mentioned, at more extreme (lower) temperatures, it may be easier to detect the occupant's skin in the image, so a wider specified range is specified. At more moderate temperatures, as mentioned, it becomes more difficult to detect occupant skin, and the specified range is narrower.

[0026] Back to Fig. ; when using the specified range, the thermographic image is compressed (i.e., converted into a temperature map) that assigns a first pixel value (e.g., a binary "1" for white) for each pixel that falls within the specified temperature range, and a second pixel value (e.g., a binary "0" for black) if a pixel corresponds to a temperature outside the specified range. A portion of the two-value temperature map is Fig. shown. Shaded rectangles 120 correspond to black pixels in the temperature map that fall outside the specified temperature range, and unshaded rectangles 121 correspond to white pixels in the thermographic image that fall within the specified range and indicate a potential target.

[0027] The following code, written for Matlab, demonstrates compressing a thermographic image into a black and white temperature map and displaying the resulting image.

[0028] %Search target areas: tol = 90; div = 30; r=252; g=243; b=102; img2a = img(:,:,1); img2b = img(:,:,2); img2c = img(:,:,3); idx1 = find(img2a>max(0,r-tol) & img2a <min(255,r+tol)); idx2 = find(img2b>max(0,g-tol) & img2b <min(255,g+tol)); idx3 = find(img2c>max(0,b-tol) & img2c <min(255,b+tol)); idx = intersect(idx1,intersect(idx2,idx3)); img2a = zeros(size(img2a)); img2b = zeros(size(img2b)); img2c = zeros(size(img2c)); img2a(idx) = 255; img2b(idx) = 255; img2c(idx) = 255; img2 = (img2a*0.2989 + img2b*0.5870 + img2c*0.1140); ax(2)=subplot(1,5,2); imshow(img2 / 255) % -------------

[0029] In step 102 of Fig. Potential target regions are merged by filtering with a sliding window. As in Fig. As shown, the sliding window preferably comprises a pixel array 122 centered on a pixel 123. During the filtering process, the window 122 is scanned over the thermographic image at each potential center pixel, with the value for each center pixel being reassigned a new value in a merged image such that the new value of that pixel corresponds to a majority value for the pixels of the temperature map in the sliding window at that position. For example, the center pixel 123 at the position shown in Fig. shown position is assigned a binary value of 0 because most of the pixels in window 122 are black. Equivalently, the average value is determined for all pixels in window 122 (e.g., by counting the white pixels and dividing by the number of pixels covered by the window), with the resulting average rounded to 0 or 1 to determine the value to be assigned to the center pixel 123. Fig. shows the merged image, with pixel 123 remaining black.

[0030] Fig. shows a subsequent position for the sliding window 122, which is centered on the next pixel 124 in the same row. Since the majority of the pixels adjacent to this new position of the sliding window 122 are black, pixel 124 is also left out of the merged image. Fig. black. When the sliding window 122 moves to the next position in the same row, a pixel 125 is evaluated, and in the example shown, the majority value remains black. Therefore, pixel 125 changes to Fig. to the value 0 (i.e., white). Thus, the filtering results in a smaller adjustment of the actual boundaries of a potential target region. On the other hand, if isolated black pixels are present in a target region, such as pixels 126 and 127, the filtering by sliding windows 122 assigns a value of 1 to these pixels based on the fact that the majority of pixels in the temperature map around these pixels are white. Thus, Fig. a merged image in which a more continuous target region is achieved by setting all pixels to white based on the use of a sliding window when the value of the pixels in the window falls, on average, within the specified temperature range.

[0031] The following Matlab code continues the example to demonstrate how the sliding window performs the filtering to stitch the target regions together.

[0032] %Merge patches to simplify: sq=20; %pad img2 w / zeros around border img3=img2; img3=[zeros(sq,size(img3,2)); img3; zeros(sq,size(img3,2))]; img3=[zeros(size(img3,1),sq) img3 zeros(size(img3,1),sq)]; for i = sq+1 : size(img3,1)-sq for j = sq+1 : size(img3,2)-sq img3(i,j) = mean(mean(img3(i-sq:i+sq,j-sq:j+sq))); % img3(i,j) = 200; end disp(i) end img3 = img3(sq:end-sq-1,sq:end-sq-1); %Remove fill img3(img3>100)=255; img3(img3<=100)=0; ax(3)=subplot(1,5,3); imshow(img3 / 255) %-------------

[0033] The sliding window comprises a square array of pixels sized to provide the best performance for removing extraneous noise while maintaining the boundaries of target regions. An array size of 20x20 pixels has been shown to be effective. Since the sliding window cannot be centered on a pixel area along the outer edges of the thermographic image, the corresponding pixels can be ignored when forming the temperature map (or additional zero-value pixels can be added around the original image to facilitate placement of the sliding window).

[0034] Back to Fig. An optional determination of seat occupancy is made in step 103 in the event that there are air outlets that can be assigned between different seats and depending on the presence or absence of occupants at the corresponding seats. Step 103 may be skipped if the steerable outlet(s) can only be directed to a specific seat. Determining seat occupancy may help determine how many air outlets are directed to each specific seat, but such a determination may alternatively be a fixed relationship or may be determined by other means, for example, by manual configuration using the HVAC HMI.

[0035] In step 104, merged regions are ranked by area. If there is only one steerable air outlet, then the ranking need only identify the largest merged region. If there are multiple steerable outlets to be used to provide airflow to a particular occupant, then the ranking identifies the appropriate number of largest merged regions to assign each individual steerable outlet to a selected area. For example, in a vehicle with two steerable air outlets serving front seat occupants and only two front seating positions (i.e., driver and front passenger), if no passenger is present, the two largest merged regions at the driver's seat would be identified and targeted.If a passenger is present on the other side, the largest merged region for each of the two seats is identified and the respective steerable outlets are directed accordingly.

[0036] Ranking the regions by area involves summing the number of white pixels in the region. The Matlab demonstration uses a function called bwboundary to sum the areas as follows. %Ranking area by size: [B,L] = bwboundaries(img3,'noholes'); for k = 1:length(B) area(k,1) = polyarea(B{k}(:,1),B{k}(:,2)); end area_sort = flipud(sortrows([area [1:length(B)]'],1)); ax(4)=subplot(1,5,4); which_areas = [1]; %Specify 1-to-N rankings for target for i=which_areas L_temp = L; L_temp(L_temp-=int16(area_sort(i,2))) = 0; L_ranked(:,:,i) = L_temp; end img4 = label2rgb(sum(L_ranked,3), @spring, [0.0 0.0 0.0]); imshow(img4) % -------------

[0037] In step 105, the centroids for the largest classified region(s) to be targeted are determined. Then, in step 106, the outlets are directed to the centroids. A preferred method for locating centroids is described in Fig. shown. In a merged image 130, a merged region 131 has been classified and identified as a target area (i.e., the region corresponds to the occupant's skin surface). A centroid 132 of the region 131 may preferably be determined by calculating an average of the orthogonal coordinates of the pixels within the region 131. Thus, the orthogonal coordinates x and y correspond to the columns and rows of image 130. The coordinates for two of the included pixels are shown as (x1,y1) and (x2,y2). The x and y values ​​for all pixels in the region 131 are averaged to determine the average coordinate for the centroid 132 (with the average value rounded down).

[0038] Further in the Matlab example, locating the centroid is demonstrated as follows. Find %centroid(s) of target area(s): for i=which_areas [row,col] = find(L_ranked(:,:,i)>0); L_ranked_cent(i,:) = [round(mean(col)) round(mean(row))]; end ax(5)=subplot(1,5,5); imshow(img4) hold on plot(L_ranked_cent(:,1),L_ranked_cent(:,2),'w.','markersize',40) plot(L_ranked_cent(:,1),L_ranked_cent(:,2),'r.','markersize',35) plot(L_ranked_cent(:,1),L_ranked_cent(:,2),'w+','markersize',10) linkaxes(ax,'x') % hold on % for k = 1:length(B) % boundary = B{k}; % plot(boundary(:,2), boundary(:,1), 'w', 'LineWidth', 2) % end KEY TO SYMBOLS

[0039] Fig. 20 HVAC system 21 electrical system 22 HVAC control 23 Infrared camera (front) 24 infrared camera (rear) 25 passenger seating area 26 Passenger seating area 27 Passenger seating area 28 Passenger seating area 30 Blower (right) 31 Controlled air vent (fitting, front, outside left) 33 Controlled air vent (fitting, front, inside left) 35 Controlled outlet (floor, left) 37 Blower (left) 38 Controlled air vent (fitting, front, inside right) 40 Controlled air vent (fitting, front, outside right) 42 Controlled outlet (floor, right) 44 Blower (rear) 45 Controlled outlet (rear left) 47 Controlled outlet (rear right) 49 Controlled outlet (floor) 51 NSchlT 52 Ambient air sensor 53 Passenger compartment air sensor 54 Passenger compartment HMI 55 seat occupancy sensors

Claims

[1] A method for controlling a heating, ventilation and air conditioning system, HVAC system, (20) having a steerable outlet (13-17) for introducing a flow of treated air into the passenger compartment (10) of a vehicle, comprising the steps of: capturing thermographic images (75) covering a fixed region in the passenger compartment (10) in which an occupant (25-28) is potentially located; Compressing a thermographic image (75) into a temperature map (76) representing pixels (123-127) of the thermographic image (75) that fall within a predetermined temperature range corresponding to the occupant (25-28); filtering the temperature map (76) according to a sliding window (122) to merge continuous regions (131) of pixels (123-127) that fall on average within the predetermined temperature range; quantifying an area for each continuous region (131); Locating a center of gravity (132) of a continuous region (131) having a largest area; and Align the steerable outlet (13-17) to the center of gravity (132). [2] The method of claim 1, further comprising the steps of: Measuring an external temperature outside the vehicle; Measuring an internal temperature inside the passenger compartment (10); and Selecting the preset temperature range in response to the external and internal temperature. [3] The method of claim 2, wherein the selecting step accesses a look-up table (51) defining a minimum temperature and a maximum temperature of the predetermined temperature range according to the measured internal and external temperatures. [4] The method of claim 1, wherein the temperature map (76) comprises pixels (123-127), each having either a first pixel value indicative of a temperature within the predetermined temperature range or a second pixel value indicative of a temperature outside the predetermined temperature range, and wherein the sliding window (122) of the filtering step comprises a square array of pixels (123-127) sequentially sampling the temperature map (76), wherein a pixel (123-127) at successive centers of the square array is assigned a value corresponding to a majority value for the pixels (123-127) of the temperature map (76) in the square array at each successive position. [5] The method of claim 1, wherein the centroid (132) is located in response to an average of orthogonal coordinates of the pixels (123-127) of the respective region (131). [6] The method of claim 1, wherein the area for each continuous region (131) is quantified in response to summing a number of pixels (123-127) in the continuous region (131). [7] Vehicle device comprising: a heating, ventilation and air conditioning system, HVAC system, (20) having a steerable outlet (13-17) for introducing a flow of treated air into the passenger compartment (10) of a vehicle; a thermographic imager configured to capture thermographic images (75) covering a fixed region in the passenger compartment (10) in which an occupant (25-28) is potentially located; and an HVAC control circuit (22) arranged to carry out a method according to one of the preceding claims. [8] The vehicle device according to claim 7, further comprising: an ambient temperature sensor that measures an external temperature outside the vehicle; and a passenger compartment temperature sensor that measures an internal temperature inside the passenger compartment (10). [9] The vehicle device of claim 7, wherein the steerable outlet (13-17) has a variable focus controllable by the HVAC control circuit (22) in response to a size of the largest area. [10] The vehicle device of claim 8, wherein the steerable outlet (13-17) has a variable flow rate that can be controlled by the HVAC control circuit (22) in response to the internal temperature.

Images