Navigation based on traffic light cycle prediction

A method for autonomous vehicle navigation, the method comprising:

• receiving from an image capture device one or more images representative of an environment of a host vehicle;
• identifying, based on analysis of the one or more images, a representation of at least one traffic light in the one or more images;
• determining a state of the at least one traffic light based on analysis of the one or more images;
• receiving from a server-based system an autonomous vehicle road navigation model, wherein the autonomous vehicle road navigation model includes stored information relating to a plurality of mapped traffic lights associated with a road segment, wherein the stored information provides one or more indicators of traffic light state timing for each of the plurality of mapped traffic lights, wherein the autonomous vehicle road navigation model includes associations between the plurality of mapped traffic lights and target trajectories along the road segment, the associations having been determined by:
• collecting navigational information from multiple vehicles traveling the road segment at different times;
• and
• determining, based on the navigational information, correlations between navigational actions of the multiple vehicles and observed state identifiers for the plurality of mapped traffic lights, wherein the navigational actions of the multiple vehicles comprise at least one of steering, braking, or accelerating, wherein the correlations between the navigational actions and the state identifiers are based on time-dependent functions, and wherein the observed state identifiers include state identifiers corresponding to traffic lights that are not positioned in front of a travel path of the host vehicle;
• determining, based on the autonomous vehicle road navigation model, at least one timing aspect associated with the determined state of the at least one traffic light state, the at least one timing aspect including an expected duration time for the determined state of the at least one traffic light;
• determining, based on analysis of the one or more images, a lane of travel associated with the host vehicle;
• determining whether the at least one traffic light is relevant to the host vehicle's trajectory based on the lane of travel and the associations included in the autonomous vehicle navigation model;
• determining, based on the expected duration time, whether the traffic state is expected to change before the host vehicle reaches the at least one traffic light;
• in response to determining that the at least one traffic light is relevant, determining at least one navigational action for the host vehicle, wherein the at least one navigational action is determined based on whether the traffic state is expected to change;
• and
• causing one or more actuator systems associated with the host vehicle to implement the determined at least one navigational action for the host vehicle.

A non-transitory computer-readable medium storing instructions that, when executed by at least one processor, are configured to cause at least one processor to perform a method, the method comprising:

• receiving from an image capture device one or more images representative of an environment of a host vehicle;
• identifying, based on analysis of the one or more images, a representation of at least one traffic light in the one or more images;
• determining a state of the at least one traffic light based on analysis of the one or more images;
• receiving from a server-based system an autonomous vehicle road navigation model, wherein the autonomous vehicle road navigation model includes stored information relating to a plurality of mapped traffic lights associated with a road segment, wherein the stored information provides one or more indicators of traffic light state timing for each of the plurality of mapped traffic lights, wherein the autonomous vehicle road navigation model includes associations between the plurality of mapped traffic lights and target trajectories along the road segment, the associations having been determined by:
• collecting navigational information from multiple vehicles traveling the road segment at different times;
• and
• determining, based on the navigational information, correlations between navigational actions of the multiple vehicles and observed state identifiers for the plurality of mapped traffic lights, wherein the navigational actions of the multiple vehicles comprise at least one of steering, braking, or accelerating, wherein the correlations between the navigational actions and the state identifiers are based on time-dependent functions, and wherein the observed state identifiers include state identifiers corresponding to traffic lights that are not positioned in front of a travel path of the host vehicle;
• determining, based on the autonomous vehicle road navigation model, at least one timing aspect associated with the determined state of the at least one traffic light state, the at least one timing aspect including an expected duration time for the determined state of the at least one traffic light;
• determining, based on analysis of the one or more images, a lane of travel associated with the host vehicle;
• determining whether the at least one traffic light is relevant to the host vehicle's trajectory based on the lane of travel and the associations included in the autonomous vehicle navigation model;
• determining, based on the expected duration time, whether the traffic state is expected to change before the host vehicle reaches the at least one traffic light;
• in response to determining that the at least one traffic light is relevant, determining at least one navigational action for the host vehicle, wherein the at least one navigational action is determined based on whether the traffic state is expected to change;
• and
• causing one or more actuator systems associated with the host vehicle to implement the determined at least one navigational action for the host vehicle.