Calculating vehicle speed for a road curve

A method for navigating a host vehicle, the method comprising:

• receiving from a camera onboard the host vehicle a captured image representative of an environment of the host vehicle;
• providing the captured image to a trained system, wherein the trained system is configured to generate an output based on a presence of a curved road segment in the captured image, wherein the trained system has undergone at least one training process based on autonomous vehicle (AV) map information, and wherein the curved road segment is associated with a road on which the host vehicle is traveling;
• receiving the output generated by the trained system, the output including at least one speed value for the host vehicle, wherein the at least one speed value output from the trained system is based on a proximity of the host vehicle to the curved road segment, based on at least one characteristic of the curved road segment represented in the captured image, and based on a maximum allowable acceleration level for the host vehicle, wherein the maximum allowable acceleration level is determined based on a detected characteristic of the environment of the host vehicle, and wherein the detected characteristic includes at least one of a highway environment or a local road environment, and a first maximum acceleration level associated with the highway environment is lower than a second maximum acceleration level associated with the local road environment;
• and
• causing the host vehicle to take at least one navigational action based on the determined at least one speed value.

A non-transitory computer-readable medium storing instructions executable by at least one processor for performing a method for navigating a host vehicle, the method comprising:

• receiving from a camera onboard the host vehicle a captured image representative of an environment of the host vehicle;
• providing the captured image to a trained system, wherein the trained system is configured to generate an output based on a presence of a curved road segment in the captured image, wherein the trained system has undergone at least one training process based on autonomous vehicle (AV) map information, and wherein the curved road segment is associated with a road on which the host vehicle is traveling;
• receiving the output generated by the trained system, the output including at least one speed value for the host vehicle, wherein the at least one speed value output from the trained system is based on a proximity of the host vehicle to the curved road segment, based on at least one characteristic of the curved road segment represented in the captured image, and based on a maximum allowable acceleration level for the host vehicle, wherein the maximum allowable acceleration level is determined based on a detected characteristic of the environment of the host vehicle, and wherein the detected characteristic includes at least one of a highway environment or a local road environment, and a first maximum acceleration level associated with the highway environment is lower than a second maximum acceleration level associated with the local road environment;
• and
• causing the host vehicle to take at least one navigational action based on the determined at least one speed value.