首页|New Findings from University of West London Describe Advances in Robotics (An Im plementation of Communication, Computing and Control Tasks for Neuromorphic Robo tics on Conventional Low- Power CPU Hardware)
New Findings from University of West London Describe Advances in Robotics (An Im plementation of Communication, Computing and Control Tasks for Neuromorphic Robo tics on Conventional Low- Power CPU Hardware)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – Researchers detail new data in robotic s. According to news originating from London, United Kingdom, by NewsRx correspo ndents, research stated, “Bioinspired approaches tend to mimic some biological f unctions for the purpose of creating more efficient and robust systems. These ca n be implemented in both software and hardware designs.” Funders for this research include University of West London. Our news journalists obtained a quote from the research from University of West London: “A neuromorphic software part can include, for example, Spiking Neural N etworks (SNNs) or event-based representations. Regarding the hardware part, we c an find different sensory systems, such as Dynamic Vision Sensors, touch sensors , and actuators, which are linked together through specific interface boards. To run real-time SNN models, specialised hardware such as SpiNNaker, Loihi, and Tr ueNorth have been implemented. However, neuromorphic computing is still in devel opment, and neuromorphic platforms are still not easily accessible to researcher s. In addition, for Neuromorphic Robotics, we often need specially designed and fabricated PCBs for communication with peripheral components and sensors. Theref ore, we developed an all-in-one neuromorphic system that emulates neuromorphic c omputing by running a Virtual Machine on a conventional low-power CPU. The Virtu al Machine includes Python and Brian2 simulation packages, which allow the runni ng of SNNs, emulating neuromorphic hardware. An additional, significant advantag e of using conventional hardware such as Raspberry Pi in comparison to purpose-b uilt neuromorphic hardware is that we can utilise the built-in physical input-ou tput (GPIO) and USB ports to directly communicate with sensors. As a proof of co ncept platform, a robotic goalkeeper has been implemented, using a Raspberry Pi 5 board and SNN model in Brian2. All the sensors, namely DVS128, with an infrare d module as the touch sensor and Futaba S9257 as the actuator, were linked to a Raspberry Pi 5 board. We show that it is possible to simulate SNNs on a conventi onal low-power CPU running real-time tasks for low-latency and low-power robotic applications.”
University of West LondonLondonUnite d KingdomEuropeEmerging TechnologiesMachine LearningRoboticsRobotsSo ftware
NETL
NSTL
