WO2018056894A1

WO2018056894A1 - Thermopile array sensor for motion tracking and its application

Info

Publication number: WO2018056894A1
Application number: PCT/SG2016/050463
Authority: WO
Inventors: Qinghua Xia; Dan GAO; Xiaowen SONG; Lanzi Jiang
Original assignee: UNITECH MECHATRONICS Pte Ltd
Current assignee: UNITECH MECHATRONICS Pte Ltd
Priority date: 2016-09-21
Filing date: 2016-09-21
Publication date: 2018-03-29
Anticipated expiration: 2019-03-21
Also published as: CN109844476A

Abstract

The system consists of thermopile infrared array sensors, an MCU implemented with motion tracking algorithms, and strategy for lighting control. A thermopile array sensor consists of a lens, a sensor chip, and an infrared image processor. The lens is for image formation. The sensor chip consists of row and columns of thermopile pixels, and placed at the focal point of the lens. The thermopile sensor use infrared radiation versus conduction for heat transfer. The infrared image processor of the sensor is used to derive an object's surface temperature captured by the pixels of the sensor, and transmit the digital temperature data to an external MCU for further processing. Motion tracking algorithms are formulated in the MCU based on the sensor's temperature measurements, for detection of moving objects, position and presence of motion and motionless objects, and objects' direction of movement. The tracking algorithms can be used for brightness control of lights to achieve energy saving.

Description

THERMOPILE ARRAY SENSOR FOR MOTION TRACKING AND ITS APPLICATION

TECHNICAL FIELD

The present disclosure describes thermopile infrared array sensor, an MCU implemented with motion tracking algorithms based on the measurements of the array sensor, and its application to lighting control system for energy saving.

BACKGROUND ART

In many places, common area lighting contributes to majority energy consumption. Traditionally, the lights at the places are turned on with full brightness at night and turned off in the morning the next day. Sometime the need for lighting is low, especially after midnight due to low human traffic. There is great potential to reduce the energy consumption by enhancing the ability of the lighting control system to provide lighting with proper brightness level only when required.

There are lighting energy saving solutions in the market that employ motion detection sensors such as passive infrared (PIR) or microwave sensors, to control brightness of lights. When no motion is detected around a motion detection sensor, the lights around the sensor are dimmed. When motion is detected around a sensor, the lights nearby are increased to full brightness. However, those motion detection sensors are unable to detect moving objects, position and presence of motionless objects, objects' direction of movement and their surface temperature.

It is therefore the objective of the invention to develop motion tracking algorithms with thermopile infrared array sensor that can detect more motion activities than PIR and microwave sensors, and apply the algorithms to control the brightness of lights in order to reduce lighting energy consumption further.

SUMMARY

According to the invention, the system consists of one or more thermopile infrared sensor arrays, an MCU, and lighting control system. A thermopile array sensor consists of a lens, a sensor chip, and an infrared image processor. The sensor chip is constructed with multiple rows and columns of thermopiles arranged in array. The thermopile sensors can be instrumented to measure the temperature of objects and people remotely.

According to the first aspect of the present invention, motion tracking algorithms are formulated from the temperature measurements of the array sensor to detect moving objects, position and presence of motion and motionless objects, and objects' direction of movement.

According to the second aspect of the present invention, the algorithms derived can be used for brightness control of lights to achieve energy saving.

BRIEF DESCRIPTION OF DRAWINGS

FIG. 1 shows construction of thermopile array sensor

FIG. 2 shows thermopile sensor array arrangement

FIG. 2 shows infrared image processor and thermopile array

FIG. 4 shows example of sensor's temperature measurement

FIG. 5 shows two thermopile array sensors mounted on ceiling for human presence detection

FIG. 6 shows measured profile of a human body temperature after data filtering

FIG. 7 shows heat map measurement of multiple persons' surface temperate from the sensor

FIG. 8 shows scenario of a person walking through a thermopile array sensor

FIG. 9 shows heat map display of a person walking towards the edge of the sensor

FIG. 10 shows lighting control strategy based on a person's movement direction

FIG. 1 1 shows different heat signatures of different persons and their moving directions

FIG. 12 shows lighting control strategy based on multiple movement directions detected DETAILED DESCRIPTION

A thermopile array sensor 101 shown in FIG. 1 consists of a lens, a sensor chip, and an infrared image processor. The lens is for image formation. The sensor chip 102 shown in FIG. 1 and FIG. 2 is constructed with thermopile pixels arranged in array, and placed at the focal point of the lens. The thermopile sensor use infrared radiation versus conduction for heat transfer. Each thermopile pixel, short for pixel, comprises of series of thermo elements, each element being a thin wire made of two materials of different thermal activity. When a temperature difference occurs between the two ends of a wire, an electrical tension develops. As illustrated in FIG. 3, the infrared image processor of the sensor 101 is used to derive temperature results of all the pixels from electrical tension measurements, and transmit the digital temperature data to an external MCU for further processing. Digital filtering is implemented to mitigate the measurement noise from the thermopiles.

FIG. 4 shows example of an object's surface temperature measurement within the thermopile sensor's field of view. Various algorithms can be implemented in an MCU based on the temperature data stream acquired.

As illustrated in FIG. 5, one or more thermopile array sensors can be mounted on ceiling for human presence detection. When a human is within the detection area of the sensor, the average temperature of all the pixels in the sensor will be higher than the average temperature when nobody is within the same detection area. Based on the temperature difference, status of human presence can be derived, regardless of motion or motionless objects in the area.

In many places for people to sit and chat, PIR or microwave sensors are not installed for light brightness adjustment, because those sensors are unable to detect motionless objects, thus the lights are kept on all the time, even if there is nobody in the area. Thermopile array sensors can be deployed there for energy saving by dimming the lights when nobody is around, and increasing the brightness of the lights when human presence is detected. Furthermore, based on the sensor's temperature data, a person's position in the area can also be estimated. When a person is under a thermopile array sensor, the measured body temperature data maybe similar to what is shown in FIG. 4, with floor and ambient temperature below 25 degree Celsius, and human body's temperature above 25 degree Celsius. Applying a filter to remove the temperature data that are below 25 degree Celsius, the heat map of the sensor's measurements is shown in FIG. 6, with temperature profile of human body shown around the centre of the heat map. Applying this result to lighting control, energy consumption of the lights can be further reduced by increasing the brightness of the lights only directly under the detected person, and dimming the rest of the lights at the area.

When multiple persons are under a sensor, after removing the temperature data below 25 degree Celsius, a heat map example of the measurements is shown in FIG. 7, with three distinct zones identified as three persons, thus their estimated positions and the number of persons within the detection area can also be obtained. The result can be applied to lighting control for energy saving.

Motion tracking algorithms based on the sensor readings can also be used for detecting direction of movement. As shown in FIG. 8, when a person moves from left to the centre of the sensor, after data filtering, the heat map of the person's surface temperature captured is similar to what is shown in FIG. 6. Based on the heat map, the centre of heat can be obtained, which is near the pixel showing 36 degree Celsius. When the person continues to move to the right, the heat map maybe similar to what is shown in FIG. 9, with centre of heat near the pixel showing 36 degree Celsius. Based on the movement trend of centre of heat from the sensor pixels, the person's direction of movement can be estimated. For lighting control as illustrated by FIG. 10, only the two lights along the person's movement direction are in full brightness, and the rest of the lights are at dimmed condition. With the control strategy, energy consumption can be further reduced. Different person may have different heat signature exhibiting different surface temperature. As shown in FIG. 1 1 , the temperature of the person at the top left portion of the heat map is higher than the person near the middle left. By capturing the movement of the persons with different temperature signature, the movement directions of the persons can be estimated, with one person walking from top left to top right, and another person walking from middle left to bottom right. Applying this result to lighting control, only the lights near the sensors and along the two moving directions are fully turned on, and the rest are tuned off or not in full brightness in order to reduce energy further, as illustrated in FIG. 12.

Claims

1 . A thermopile array sensor comprising:

a lens for image formation, a sensor chip with rows and columns of thermopile pixels placed at the focal point of the lens, and an infrared image processor to derive temperature data from the thermopiles' measurements, and transmit the filtered digital temperature data to an external MCU for motion tracking algorithms formulation and further processing.

2. For the thermopile array sensor of claim 1 , for human presence detection purpose, when a human is within the detection area of the sensor, the average temperature of all the thermopile pixels in the sensor will be higher than the average temperature when nobody is within the same detection area, and based on the temperature difference, status of human presence can be derived with an MCU, regardless of motion or motionless objects in the area.

3. For the human presence detection algorithms of claim 2, thermopile array sensors can be deployed at places where people sit and chat, with implemented energy saving strategy of dimming the lights when nobody is around, and increasing the light brightness when human presence is detected.

4. For the thermopile array sensor of claim 1 , when a person is within thermopile array sensor's detection area, the measured body temperature maybe higher than the ambient temperature, and by filtering out the ambient temperature data, centre of heat, thus the rough position of a human body within the area can be estimated.

5. For the human position detection algorithm of claim 4, applying the result to lighting control, energy consumption of the lights can be reduced by increasing the brightness of the lights that are located directly under the detected person, and dimming the rest of the lights.

6. For thermopile array sensor of claim 1 , when multiple persons are within the detection area of the sensor, different person may have different heat signature exhibiting different average body temperature, and by calculating each person's centre of heat, the rough positions of the persons and number of persons can be obtained.

7. For the multiple human positions detection algorithm of claim 6, applying the result to lighting control, energy consumption of the lights can be reduced by increasing the brightness of the lights directly under each detected person, and dimming the rest of the lights without human presence nearby.

8. For thermopile array sensor of claim 1 , when a person walks within the detection area of the sensor, the positions of the person's centre of heat at different time stamps can be calculated, and with the estimated positions, the direction of movement of the person can be derived.

9. For the motion tracking algorithm of claim 8, applying this result to lighting control, only the lights near the person and along the person's moving direction are fully turned on, and the rest of the lights are turned off or not in full brightness in order to reduce energy consumption.

10. For thermopile array sensor of claim 1 , when multiple persons walk within the detection area of the sensor, because different person may exhibit different average body temperature, and by calculating the positions of each person's centre of heat at different time stamps, their estimated directions of movement can be derived.

1 1 . For the motion tracking algorithm of claim 10, applying this result to lighting control, only the lights near and along each person's moving direction are fully turned on, and the rest are turned off or not in full brightness in order to reduce energy consumption.