Edge AI is when your results are calculated on or near the device (such as a camera) that captures the data you are analysing, either because you need answers in real-time, or because sending data to the cloud for analysis will be too costly.
For example, edge AI could be used on an autonomous delivery robot to help it understand and react to the world around it. A device like this will always need real-time results to ensure the safety of others, and with so many sensors onboard that create masses of data, it will be too costly to send the data to the cloud for analysis.
• Fruit and vegetable picking robots.
• Automated weeding robots.
• Individual animal identification and monitoring.
• Fruit/veg quality grading and sorting.
• Automated pick and pack robots.
• Last mile delivery robots.
• Package sorting and routing.
• Object and threat detection.
• Facial recognition.
• People counting.
• Mask detection and social distancing.
• Number plate recognition.
Development kits to help test and prototype your edge AI solution are readily available, but are rarely suitable for use in the environment where they will ultimately be deployed.
A purpose-built industrial-grade AI system based on Nvidia Jetson, Intel Movidius, or a high-end GPGPU, will be more appropriate and will offer more reliable long-term operation.
Our edge AI systems are designed for use in environments where shock, vibration, or extreme temperatures can be a problem, and will run 24/7 for many years.
For ultra-complex AI workloads, such as autonomous driving, combining an industrial computer with a high-performance graphics card will give you everything you need for real-time results in harsh environments. Further information