AVM - METHOD AND DEVICE FOR ATTENTION-DRIVEN RESOURCE ALLOCATION BY USING AVM AND REINFORCEMENT LEARNING TO THEREBY ACHIEVE SAFETY OF AUTONOMOUS DRIVING

摘要

The present invention is a method for achieving better autonomous driving performance while saving computing power by using a confidence score that represents the credibility of object detection, generated in parallel with the object detection process, (a) obtaining, by a computing device, at least one circumstance image of the surroundings of the target vehicle through at least one panorama view sensor installed in the target vehicle; (b) causing, by the computing device, a Convolutional Neural Network (CNN) to apply a CNN operation to the context image at least once to generate initial object information and initial confidence information for the context image; And (c) generating, by the computing device, with the support of the reinforcement learning agent, generating final object information for the context image with reference to initial object information and initial confidence information.