Systems and methods for pedestrian crossing risk assessment and directional warning

A computer-implemented system comprising:

• one or more sensors of a vehicle, the one or more sensors to receive data indicative of a pedestrian in an environment corresponding to the vehicle, the data including trajectory data corresponding to the pedestrian;
• and
• parallel processing circuitry configured to:
• determine one or more attributes of the pedestrian based at least in part on the data indicative of the pedestrian;
• calculate a risk level representing a likelihood of a collision associated with the pedestrian based on the trajectory data and on the one or more attributes, wherein the calculation is performed at least in part using a machine learning model;
• and
• cause an operation to be performed in response to the risk level, wherein the operation includes providing a warning directed to the pedestrian.

A non-transitory computer readable medium having instructions encoded thereon, that when executed by processing circuitry causes the processing circuitry to:

• receive, from one or more sensors of a vehicle, data indicative of a trajectory of a pedestrian in an environment corresponding to the vehicle;
• determine one or more attributes of the pedestrian;
• calculate, using a machine learning model, a risk level representing a likelihood of a collision associated with the pedestrian based on the data indicative of the trajectory and on the one or more attributes;
• and
• cause an operation to be performed in response to the risk level to reduce the risk level, wherein the operation includes providing a warning directed to the pedestrian.