Technology, Artificial IntelligenceAugust 1, 2023Cutting-Edge innovation enables machines to see in darkness like daylight, revolutionizing perception and roboticsResearchers at Purdue University have developed a patent-pending method called HADAR (heat-assisted detection and ranging) that enhances machine vision and perception. This innovation combines thermal physics, infrared imaging, and machine learning to enable machines to perceive their surroundings in pitch darkness, similar to broad daylight. Traditional methods like LiDAR, radar, and sonar have limitations, while thermal imaging, although capable of sensing through darkness and adverse weather, lacks texture and features. HADAR overcomes these challenges by recovering texture and accurately identifying temperature, emissivity, and texture of objects in a scene. The team aims to improve the size and speed of data collection for wider applications in autonomous vehicles, robots, agriculture, defense, healthcare, and wildlife monitoring. The technology has been disclosed to the Purdue Innovates Office of Technology Commercialization, with funding from DARPA to support further research and development.
Source: Purdue University