能承受4万次以上弯折！清华大学获柔性芯片重要突破

Core Insights - The research team from Tsinghua University's School of Integrated Circuits has developed the "FLEXI flexible digital storage-computing chip," marking a significant breakthrough in flexible electronics and edge AI hardware in China, filling a technological gap in high-performance flexible AI computing chips [1][3] Group 1: Technology and Innovation - The newly developed flexible AI chip utilizes CMOS low-temperature polycrystalline silicon (LTPS) technology, allowing direct manufacturing on flexible substrates, offering advantages of low power consumption, low cost, and high integration [3] - The research team innovatively increased the number of metal layers, overcoming the traditional bottleneck of complex chip interconnections in flexible electronics [3] - The chip employs a digital "in-memory computing" architecture, processing data within the memory, which eliminates the time and energy costs associated with data transfer and surpasses the performance limitations of the "memory wall," outperforming traditional analog solutions [3] Group 2: Performance and Applications - Test data shows that the chip can operate stably in folded and curled states, maintaining its computational capability after 40,000 cycles of folding, and exhibits good resistance to temperature, humidity, and light aging [3] - The minimum manufacturing cost of the chip is only $0.016, making it suitable for integration into wearable devices to recognize daily human activities using physiological signals such as heart rate, respiratory rate, and body temperature [3] Group 3: Future Prospects - Experts have noted that this technology fills a gap in AI-specific computing hardware within the flexible electronics sector [3] - Future enhancements through the application of new semiconductor materials and power gating technology optimization are expected to further improve performance [3] - Continuous optimization of production yield and chip size could drive industrial upgrades and technological innovations in wearable health devices and IoT terminals [3]