Environment controller and method for inferring one or more commands for controlling an appliance taking into account room characteristics
Method and environment controller for inferring via a neural network one or more commands for controlling an appliance. A predictive model generated by a neural network training engine is stored by the environment controller. The environment controller determines at least one room characteristic. The environment controller receives at least one environmental characteristic value and at least one set point. The environment controller executes a neural network inference engine, which uses the predictive model for inferring the one or more commands for controlling the appliance. The inference is based on the at least one environmental characteristic value, the at least one set point and the at least one room characteristic. The environment controller transmits the one or more commands to the controlled appliance.
1. An environment controller, comprising:
a communication interface;
memory for storing a predictive model generated by a neural network training engine, the predictive model comprising weights of a neural network determined by the neural network training engine; and
a processing unit comprising at least one processor for:
receiving at least one room characteristic via one of the communication interface and a user interface of the environment controller, and storing the at least one room characteristic in the memory;
receiving via the communication interface a current room occupancy;
receiving via one of the communication interface and the user interface a target temperature;
executing a neural network inference engine, the neural network inference engine implementing a neural network using the predictive model for inferring an output based on inputs, the output comprising one or more command for controlling an appliance, the inputs comprising the at least one room characteristic, the current room occupancy and the target temperature; and
transmitting the one or more command to the controlled appliance via the communication interface;
wherein the at least one room characteristic used as input of the neural network comprises at least one of the following: a room type identifier selected among a plurality of room type identifiers, one or more geometric characteristics of the room, and a human activity in the room.
2. The environment controller of claim 1 , wherein the at least one room characteristic comprises one or more geometric characteristics of the room, the one or more geometric characteristics of the room comprising at least one of the following: a volume of the room, a surface of the room, a height of the room, a length of the room, and a width of the room.
3. The environment controller of claim 1 , wherein the at least one room characteristic comprises a human activity in the room, the human activity in the room comprising at least one of the following: periods of time when the room is occupied by humans, and a type of activity performed by humans occupying the room.
4. The environment controller of claim 1 , wherein the processing unit further receives via the communication interface at least one additional environmental characteristic value; and wherein the inputs of the neural network inference engine further comprise the at least one additional environmental characteristic value.
5. The environment controller of claim 4 , wherein the at least one additional environmental characteristic value comprises at least one of the following: a current temperature, a current humidity level, and a current carbon dioxide (CO2) level.
6. The environment controller of claim 1 , wherein the current room occupancy consists of a determination whether the room is occupied or not, or the current room occupancy consists of a number of persons present in the room.
7. The environment controller of claim 1 , wherein the processing unit further receives via one of the communication interface and the user interface at least one additional set point; and wherein the inputs of the neural network inference engine further comprise the at least one additional set point.
8. The environment controller of claim 7 , wherein the at least one additional set point comprises at least one of the following: a target CO2 level, and a target humidity level.
9. The environment controller of claim 1 , wherein the one or more command for controlling the appliance includes at least one of the following: a command for controlling a speed of a fan, a command for controlling a pressure generated by a compressor, and a command for controlling a rate of an airflow through a valve.
10. The environment controller of claim 1 , wherein the controlled appliance consists of a heating, ventilation, and air-conditioning (HVAC) appliance.
11. A method for inferring via a neural network one or more command for controlling an appliance for a room, the method comprising:
storing a predictive model generated by a neural network training engine in a memory of an environment controller, the predictive model comprising weights of a neural network determined by the neural network training engine;
determining by a processing unit of the environment controller at least one room characteristic;
receiving by the processing unit via a communication interface of the environment controller a current room occupancy;
receiving by the processing unit via one of the communication interface and a user interface of the environment controller a target temperature;
executing by the processing unit a neural network inference engine, the neural network inference engine implementing a neural network using the predictive model for inferring an output based on inputs, the output comprising the one or more command for controlling the appliance, the inputs comprising the at least one room characteristic, the current room occupancy and the target temperature; and
transmitting by the processing unit the one or more command to the controlled appliance via the communication interface;
wherein the at least one room characteristic used as input of the neural network comprises at least one of the following: a room type identifier selected among a plurality of room type identifiers, one or more geometric characteristics of the room, and a human activity in the room.
12. The method of claim 11 , wherein the at least one room characteristic comprises one or more geometric characteristics of the room, the one or more geometric characteristics of the room comprising at least one of the following: a volume of the room, a surface of the room, a height of the room, a length of the room, and a width of the room.
13. The method of claim 11 , wherein the at least one room characteristic comprises a human activity in the room, the human activity in the room comprising at least one of the following: periods of time when the room is occupied by humans, and a type of activity performed by humans occupying the room.
14. The method of claim 11 , further comprising receiving by the processing unit via the communication interface at least one additional environmental characteristic value; and wherein the inputs of the neural network inference engine further comprise the at least one additional environmental characteristic value.
15. The method of claim 14 , wherein the at least one additional environmental characteristic value comprises at least one of the following: a current temperature, a current humidity level, and a current carbon dioxide (CO2) level.
16. The method of claim 11 , wherein the current room occupancy consists of a determination whether the room is occupied or not, or the current room occupancy consists of a number of persons present in the room.
17. The method of claim 11 , further comprising receiving by the processing unit via one of the communication interface and the user interface at least one additional set point; and wherein the inputs of the neural network inference engine further comprise the at least one additional set point.
18. The method of claim 17 , wherein the at least one additional set point comprises at least one of the following: a target CO2 level, and a target humidity level.
19. The method of claim 11 , wherein the one or more command for controlling the appliance includes at least one of the following: a command for controlling a speed of a fan, a command for controlling a pressure generated by a compressor, and a command for controlling a rate of an airflow through a valve.
20. The method of claim 11 , wherein the controlled appliance consists of a heating, ventilation, and air-conditioning (HVAC) appliance.
21. A non-transitory computer program product comprising instructions executable by a processing unit of an environment controller, the execution of the instructions by the processing unit of the environment controller providing for inferring via a neural network one or more command for controlling an appliance for a room by:
storing a predictive model generated by a neural network training engine in a memory of the environment controller, the predictive model comprising weights of a neural network determined by the neural network training engine;
determining by the processing unit at least one room characteristic;
receiving by the processing unit via a communication interface of the environment controller a current room occupancy;
receiving by the processing unit via one of the communication interface and a user interface of the environment controller a target temperature;
executing by the processing unit a neural network inference engine, the neural network inference engine implementing a neural network using the predictive model for inferring an output based on inputs, the output comprising the one or more command for controlling the appliance, the inputs comprising the at least one room characteristic, the current room occupancy and the target temperature; and
transmitting by the processing unit the one or more command to the controlled appliance via the communication interface;
wherein the at least one room characteristic used as input of the neural network comprises at least one of the following: a room type identifier selected among a plurality of room type identifiers, one or more geometric characteristics of the room, and a human activity in the room.