Media coverage
4
Media coverage
Title Animal brain inspired AI game changer for autonomous robots Degree of recognition Regional Media name/outlet TU Delft Media type Web Duration/Length/Size 1 pag. Country/Territory Netherlands Date 15/05/24 Description A team of researchers at Delft University of Technology has developed a drone that flies autonomously using neuromorphic image processing and control based on the workings of animal brains. Animal brains use less data and energy compared to current deep neural networks running on GPUs (graphic chips). Neuromorphic processors are therefore very suitable for small drones because they don’t need heavy and large hardware and batteries. The results are extraordinary: during flight the drone’s deep neural network processes data up to 64 times faster and consumes three times less energy than when running on a GPU. Further developments of this technology may enable the leap for drones to become as small, agile, and smart as flying insects or birds. The findings were recently published in Science Robotics. Producer/Author Marc de Kool URL https://www.tudelft.nl/en/2024/tu-delft/animal-brain-inspired-ai-game-changer-for-autonomous-robots Persons G.C.H.E. de Croon, J.J. Hagenaars, F. Paredes Valles, S. Stroobants Title Op hersenen geinspireerde AI game-changer voor autonome robots Degree of recognition National Media name/outlet TU Delft Media type Web Duration/Length/Size 1 pag. Country/Territory Netherlands Date 15/05/24 Description Een team van onderzoekers van de TU Delft heeft een drone ontwikkeld die autonoom kan vliegen met behulp van neuromorfe beeldbewerking en besturing die zijn gebaseerd op de werking van dierenhersenen. Dierenhersenen gebruiken minder energie dan de huidige diepe neurale netwerken die op GPU’s (grafische chips) draaien. Neuromorfe processoren zijn daarom zeer geschikt voor kleine drones omdat er geen zware en grote hardware en batterijen voor nodig zijn. De resultaten zijn veelbelovend: tijdens de vlucht verwerkt het diepe neurale netwerk van de drone op neuromorfe technologie data tot 64 keer sneller en verbruikt het drie keer minder energie dan bij gebruik van een GPU. Verdere ontwikkelingen van deze technologie kunnen de sprong mogelijk maken voor drones om net zo klein, wendbaar en slim te worden als vliegende insecten of vogels. De bevindingen zijn onlangs gepubliceerd in Science Robotics. Producer/Author Marc de Kool URL https://www.tudelft.nl/2024/tu-delft/op-hersenen-geinspireerde-ai-game-changer-voor-autonome-robots Persons F. Paredes Valles, J.J. Hagenaars, S. Stroobants, G.C.H.E. de Croon Title Animal brain inspired AI game changer for autonomous robots Degree of recognition International Media name/outlet My Science Media type Web Duration/Length/Size 1 pag. Country/Territory Netherlands Date 15/05/24 Description A team of researchers at Delft University of Technology has developed a drone that flies autonomously using neuromorphic image processing and control based on the workings of animal brains. Animal brains use less data and energy compared to current deep neural networks running on GPUs (graphic chips). Neuromorphic processors are therefore very suitable for small drones because they don’t need heavy and large hardware and batteries. The results are extraordinary: during flight the drone’s deep neural network processes data up to 64 times faster and consumes three times less energy than when running on a GPU. Further developments of this technology may enable the leap for drones to become as small, agile, and smart as flying insects or birds. The findings were recently published in Science Robotics. URL https://www.myscience.org/en/news/2024/animal_brain_inspired_ai_game_changer_for_autonomous_robots-2024-tudelft Persons F. Paredes Valles, G.C.H.E. de Croon, J.J. Hagenaars, S. Stroobants Title New neural tech could power insect-sized intelligent flying robots Degree of recognition International Media name/outlet Interestingengineering Media type Web Duration/Length/Size 1 pag. Date 15/05/24 Description The system uses a five-layer spiking neural network with 28,800 neurons to analyze raw event-based camera data and estimate the camera’s 3D motion. Producer/Author Jijo Malayil URL https://interestingengineering.com/innovation/neuromorphic-technology-drone-flight Persons F. Paredes Valles, J.J. Hagenaars, S. Stroobants, G.C.H.E. de Croon