Redundant lane detection for autonomous vehicles

At least one non-transitory computer-readable storage media having computer-executable instructions embodied thereon for controlling navigation of an autonomous vehicle, wherein when executed, the computer-executable instructions cause a system to:

• receive first sensor data from at least one first sensor of the autonomous vehicle and second sensor data from at least one second sensor of the autonomous vehicle, the at least one first sensor being of a first modality, the at least one second sensor being of a second modality;
• generate a first prediction based on the first sensor data;
• generate a second prediction based on the second sensor data;
• compute a confidence value for the second prediction based on the first prediction and the second prediction;
• and
• navigate the autonomous vehicle based on the confidence value.

A computer-implemented method for controlling navigation of an autonomous vehicle, the method comprising:

• receiving, by at least one processor, first sensor data from at least one first sensor of the autonomous vehicle and second sensor data from at least one second sensor of the autonomous vehicle, the at least one first sensor being of a first modality, the at least one second sensor being of a second modality;
• generating, by the at least one processor, a first prediction based on the first sensor data using a first modality;
• generating, by the at least one processor, a second prediction based on the second sensor data using a second modality;
• computing, by the at least one processor, a confidence value for the second prediction based on the first prediction and the second prediction;
• and
• navigating, by the at least one processor, the autonomous vehicle based on the confidence value.