Inference server and environment controller for inferring one or more commands for controlling an appliance taking into account room characteristics

A method for inferring via a neural network one or more command for controlling an appliance, the method comprising:

• storing in a memory of an inference server 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;
• determining by a processing unit of an environment controller at least one room characteristic;
• receiving by the processing unit of the environment controller via a communication interface of the environment controller a current temperature and a current humidity level;
• receiving by the processing unit of the environment controller via one of the communication interface of the environment controller and a user interface of the environment controller a target temperature;
• transmitting by the processing unit the current temperature, the current humidity level, the target temperature and the at least one room characteristic to the inference server via the communication interface of the environment controller;
• receiving by a processing unit of the inference server the current temperature, the current humidity level, the target temperature and the at least one room characteristic from the environment controller via a communication interface of the inference server;
• executing by the processing unit of the inference server 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 current temperature, the current humidity level, the target temperature, and the at least one room characteristic;
• transmitting by the processing unit of the inference server the one or more command for controlling the appliance inferred by the neural network inference engine to the environment controller via the communication interface of the inference engine;
• receiving by the processing unit of the environment controller the one or more command for controlling the appliance from the inference server via the communication interface of the environment controller;
• and
• transmitting by the processing unit of the environment controller the one or more command to the controlled appliance via the communication interface of the environment controller;
• 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.