Trajectory prediction with data normalization

A system comprising:

• one or more processors coupled to a memory comprising non-transitory computer instructions that when executed by the one or more processors perform operations comprising:
• receiving, from a client device, a plurality of observed speed points, each of the plurality of observed speed points corresponding to a different slice of a plurality of slices of an observed trajectory;
• retrieving a first timestamp of a first of the plurality of observed speed points representing when the first of the plurality of speed points was received from the client device and a second timestamp of a second of the plurality of observed speed points representing when the second of the plurality of speed points was received from the client device, the first and second speed points being determined by the client device using a wireless communication antenna;
• determining a sampling rate of the client device based on a difference between the first timestamp and the second timestamp;
• processing the plurality of observed speed points corresponding to the observed trajectory by a machine learning technique to generate a plurality of predicted speed points, the machine learning technique being trained to establish a relationship between a plurality of training observed speed points and training predicted speed points;
• determining a future trajectory based on the plurality of predicted speed points, each of the plurality of predicted speed points corresponding to a different slice of a plurality of slices of the future trajectory and based on the sampling rate;
• determining that a target access control device is within a threshold range of the future trajectory;
• and
• in response to determining that the target access control device is within the threshold range of the future trajectory, performing an operation associated with the target access control device.

A non-transitory computer readable medium comprising non-transitory computer-readable instructions for performing operations comprising:

• receiving, from a client device, a plurality of observed speed points, each of the plurality of observed speed points corresponding to a different slice of a plurality of slices of an observed trajectory;
• retrieving a first timestamp of a first of the plurality of observed speed points representing when the first of the plurality of speed points was received from the client device and a second timestamp of a second of the plurality of observed speed points representing when the second of the plurality of speed points was received from the client device, the first and second speed points being determined by the client device using a wireless communication antenna;
• determining a sampling rate of the client device based on a difference between the first timestamp and the second timestamp;
• processing the plurality of observed speed points corresponding to the observed trajectory by a machine learning technique to generate a plurality of predicted speed points, the machine learning technique being trained to establish a relationship between a plurality of training observed speed points and training predicted speed points;
• determining a future trajectory based on the plurality of predicted speed points, each of the plurality of predicted speed points corresponding to a different slice of a plurality of slices of the future trajectory and based on the sampling rate;
• determining that a target access control device is within a threshold range of the future trajectory;
• and
• in response to determining that the target access control device is within the threshold range of the future trajectory, performing an operation associated with the target access control device.