Collision avoidance system

A vehicle control system, comprising:

• a processor;
• and
• a memory coupled to the processor to store instructions, which when executed by the processor, cause the processor to perform operations, the operations comprising:
• obtaining sensor data from one or more sensors, the sensor data comprising navigation characteristics of an ego vehicle and an obstacle;
• inferring one or more categories of the obstacle based on the obtained sensor data;
• based on the inferred one or more categories, selecting one or more prediction models, wherein at least one of the one or more prediction models outputs a reachability parameter corresponding to a navigation characteristic input, wherein the reachability parameter is indicative of any possibility of a collision between the ego vehicle and the obstacle;
• determining any possibility of a collision between the ego vehicle and the obstacle based on the one or more selected prediction models and the navigation characteristics;
• and
• based on the determination of any possibility of a collision, selectively performing one or more actions to avoid or mitigate a possible collision.

A non-transitory machine-readable medium having instructions stored therein, which when executed by a processor, cause the processor to perform operations, the operations comprising:

• obtaining sensor data from one or more sensors, the sensor data comprising navigation characteristics of an ego vehicle and an obstacle;
• inferring one or more categories of the obstacle based on the obtained sensor data;
• based on the inferred one or more categories, selecting one or more prediction models, wherein each of the one or more prediction models output reachability parameters corresponding to different navigation characteristic inputs, wherein each of the reachability parameters are indicative of any possibility of a collision between the ego vehicle and the obstacle;
• determining any possibility of a collision between the ego vehicle and the obstacle based on the one or more selected prediction models and the navigation characteristics;
• and
• based on the determination of any possibility of a collision, selectively performing one or more actions to avoid or mitigate a possible collision.