Startup raises €5m for neuromorphic edge chip
Innatera Nanosystems是荷兰代尔夫特技术大学的衍生产品，并开发了一种神经形态芯片，该芯片模仿了大脑的模式识别机制。这使得传感器数据的处理速度比传统处理器更快，最多可减少500倍的能量。
Target applications include intelligent speech processing in human-machine interfaces, vitals monitoring in wearable devices, target recognition in Radars and Lidars, and fault detection in industrial and automotive equipment.
该技术依赖于新的模拟信号设备，从而重现大脑基本构件的行为 - 尖刺神经元和突触。用尖峰神经元建造的神经网络具有精确的时间概念，使它们比传统的人工神经网络更加紧凑，尤其是对于涉及具有高空间和时间相关性数据的应用。
The €5m seed investment round was led by Munich-based fund MIG Verwaltungs and the Industrial Technologies Fund of venture capital group btov Partners.
Innatera is working on a suite of proprietary algorithms and an extensive software toolchain for developers to use the neuromorphic silicon. The investment will enable the company to scale up its R&D efforts and accelerate product development to deliver on customer commitments through 2021 as well as hire more staff.
“Innatera sets itself apart from the plethora of AI accelerator companies by focussing on the edge of the edge – sensory data processing in the field,” said Dr Christian Reitberger, Partner at btov. “Translating truly brain inspired design principles into state of the art analog-mixed signal solutions enables a performance envelope not accessible to more conventional solutions.”
Edge computing is gaining traction across domains including consumer electronics, IoT, smart industry, and automotive. IDC forecasts the edge AI processor market to reach the $40bn revenue mark by 2023. The market is expected to grow with a CAGR of over 85 percent as vendors integrate more AI-driven functionalities closer to sensors due to requirements for lower latency, communication cost constraints, and concerns over data privacy.
A significant portion of the projected growth in this space is expected to come from always-on sensing applications that involve continuous monitoring of sensor data, and thus require energy-efficient processor chips that fit within the device. This is where Innatera finds that conventional processors and even other emerging accelerators fall short.
“当今最有影响力的传感器驱动的应用程序受处理器的效率和速度的限制，在小型，电池供电的设备中，它比其他任何地方都更加如此。Innatera通过将模拟信号神经形态硅的能源效率与真正的尖峰神经网络算法的性能提升相结合，从而重新发明传感器的处理。”
Innatera says it has been working with large international companies in the consumer, industrial, and automotive markets.
Related articles
- OPTICAL FLOW AI STARTUP OPTERAN RAISES £2.1m
- FIRST EVENT-DRIVEN AI CAMERA SHIPS
- KIT PROVIDES ALGORITHMS FOR EVENT-DRIVEN IMAGE SENSOR
- UNDERSTANDING THE PERFORMANCE OF EVENT-BASED VISION SENSORS
Other articles on eeNews Europe
