Classification systems and methods for in-vehicle sensing with radar

A method for in-vehicle sensing and classification via frequency modulated continuous wave (FMCW) radar, the method comprising:

• transmitting, via a FMCW radar sensor, radar transmission signals;
• receiving, via the FMCW radar sensor, radar reflection signals based on the radar transmission signals, the radar reflection signals including a plurality of chirps across a plurality of frames;
• generating classification data based on the radar reflection signals to determine a sensing state inside a vehicle, the classification data including class data that classifies a radar subject;
• and
• generating a system response to provide an action concerning the sensing state inside the vehicle based on the classification data,
• wherein,
• the system response includes generating an alert when the class data classifying the radar subject is indicative of an animate subject being inside the vehicle, and
• the class data and the system response are generated when the vehicle has been parked for a predetermined time period.

A system comprising:

• a transceiver module configured to operate with frequency modulated continuous wave (FMCW) radar, the transceiver module being located inside a vehicle and being operable to (i) transmit radar transmission signals, (ii) receive radar reflection signals based on the radar transmission signals, the radar reflection signals including a plurality of chirps across a plurality of frames, and (iii) convert the radar reflection signals into digital signals;
• and
• a processor in data communication with the transceiver module, the processor being operable to perform a method;
• and
• a non-transitory computer readable medium in data communication with the processor, the non-transitory computer readable medium including instructions stored thereon that, when executed by the processor, causes the processor to perform the method that includes
• generating first range data by performing a fast Fourier transform (FFT) on a first set of chirps of a first frame using the digital signals;
• generating second range data by performing the FFT on a second set of chirps of a second frame using the digital signals;
• generating Doppler data by performing the FFT on at least the first range data and the second range data;
• generating point cloud data of a radar subject using the Doppler data, the point cloud data including location data of the radar subject;
• extracting Doppler features from the Doppler data, the Doppler features including a velocity of the radar subject;
• concatenating the point cloud data and the Doppler features as a multi-dimensional matrix,
• generating, via a classifier, classification data based on the point cloud data and the Doppler features to determine a sensing state inside the vehicle, the classification data including class data that classifies the radar subject;
• and
• generating a system response to provide an action concerning the sensing state of an interior region of the vehicle based on the classification data,
• wherein,
• the classifier generates the classification data as output data upon receiving the multi-dimensional matrix as input data, and
• the classifier includes a machine learning model or a classification model, the classification model performing classification based on a set of rules.