Smart low-power anti-collision system for drones (and cars)ДЄ key application for the new imec chip is a low-latency, low-power anti-collision system for drones. Contrary to analog SNN implementations, imecвЂ™s event-driven digital design makes the chip behave exactly and repeatedly as predicted by the neural network simulation tools. Its generic architecture based on a completely new digital hardware design means it can also easily be reconfigured to process a variety of other sensory inputs like sonar, radar and lidar data. Imec said its chip was initially designed to support electrocardiogram (ECG) and speech processing in power-constrained devices. The technology we are introducing today is a major leap forward in the development of truly self-learning systems.вЂќ WhatвЂ™s more, the spiking neurons on our chip can be connected recurrently вЂ“ turning the SNN into a dynamic system that learns and remembers temporal patterns. As such, energy consumption can significantly be reduced. вЂњSNNs operate very similarly to biological neural networks, in which neurons fire electrical pulses sparsely over time, and only when the sensory input changes. вЂњToday, we present the worldвЂ™s first chip that processes radar signals using a recurrent spiking neural network,вЂќ says Ilja Ocket, program manager of neuromorphic sensing at imec. ThatвЂ™s where spiking neural networks (SNNs) can help. Additionally, ANNsвЂ™ underlying architecture and data formatting requires data to undertake a time-consuming journey from the sensor device to the AI inference algorithm before a decision can be made.

For one, they consume too much power to be integrated into increasingly constrained (sensor) devices. They are a key ingredient, for instance, of the radar-based anti-collision systems commonly used in the automotive industry. While the chipвЂ™s architecture and algorithms can easily be tuned to process a variety of sensor data (including electrocardiogram, speech, sonar, radar and lidar streams), its first use-case will encompass the creation of a low-power, highly intelligent anti-collision radar system for drones that can react much more effectively to approaching objects.ДЄrtificial neural networks (ANNs) have already been established in a wide range of application domains. For example, micro-Doppler radar signatures can be classified using only 30 mW of power. Mimicking the way groups of biological neurons operate to recognize temporal patterns, imec said its chip consumes 100 times less power than traditional implementations while featuring a tenfold reduction in latency вЂ“ enabling almost instantaneous decision-making. Imec claims to have built the worldвЂ™s first spiking neural network (SNN) based chip for radar signal processing, enabling the creation of applications such as smart, low-power anti-collision radar systems for drones that identify approaching objects in a matter of milliseconds

Neural network radar