According to reports, aistorm, an innovative developer of AI in sensor processors with high performance, recently announced the completion of a round of financing of US $13.2 million, with investors including egis, a major supplier of biometric technology for smart phones, games and advanced driving assistance systems (ADAS) Technology); tower jazz, a global professional foundry leader; Meyer Corporation, a global leader in food manufacturing equipment; and linear dimensions semiconductor, a leader in biometrics and digital health products.
“This round of financing will help us accelerate engineering development and marketing and push innovative machine learning technologies to the edge of the Internet of things,” said David schie, chief executive of aistorm. Aistorm’s revolutionary solution can realize edge intelligent solution in low-cost simulation technology. Compared with GPU solution, aistorm solution can save 5 to 10 times of cost under the premise of guaranteed performance. “
Direct use of sensor data, without digitization, real-time processing of edge devices
The semiconductor industry is working hard to process data in edge devices to reduce the cost and security risk of transmitting large amounts of raw data from edge sensors. Artificial intelligence system requires data before information processing to be provided in digital format, but sensors capture analog data.
Processing these digital information requires advanced high-cost GPU, which is not ideal for mobile devices. They need continuous digital data input, which will consume a lot of power and introduce inevitable digital delay. Aistorm aims to solve these problems by directly processing sensor data in raw analog format in real time.
Aistorm technology is analog and mixed signal technology, but more than that. Aistorm technology enables the sensor itself to intelligently process events and make decisions for the first time, and can perform complex analysis before other schemes receive digital sensor data for processing.
This can achieve a wide range of functions and applications that can not be achieved in the past. For example, smart cars can brake faster; cameras that only run for several years with a single battery; biometric login through small wearable devices such as smart watches, smart bracelets and smart glasses; hand gesture control and basic control of handheld devices through screen image sensors and advanced machine vision It can be applied to the safety of heart.
“The response time saved by aistorm’s solution means that a more advanced driver assistance system can detect objects more quickly, achieve safe parking more intelligently, and avoid fatal collisions,” said Russell Ellwanger, chief executive of tower jazz.
“Edge applications need to process a large amount of data captured by sensors, and the digital processing of these data takes time, which means that these applications have no time to intelligently select data from the sensor data stream, and can only passively collect a large amount of data and process it later. Aistorm’s solution enables us to simplify data from sensor data stream in real time and intelligently for the first time, and handle a large number of sensor input tasks in a timely manner, “said Todd Lin, chief operating officer of aegis.
“It is very meaningful to combine artificial intelligence processing with image sensor to eliminate the high-cost digital process. For our customers, this will open up new possibilities for intelligent, event driven applications, as well as high-speed processing of edge devices, “said Dr. avi strum, senior vice president and general manager of sensor business at tower jazz.
Editor in charge: PJ