HW/SW Co-Design for Integrated AI Systems: Challenges, Use Cases and Steps Ahead

Artificial Intelligence (AI) applications are permeating large parts of the economy, science and society. Enabled by recent advances in algorithms, computer architectures and big data, AI has made significant breakthroughs in a wide range of applications. Not only in the field of computer vision, speech recognition and processing, but also in robotics and many other areas. The trend in many domains is to shift intelligence from the cloud to the edge. However, integrating AI systems in the edge requires the development of powerful solutions under very strict resource constraints. Often, there already exist solutions delivering very good results in the high-performance area, but the use of AI approaches in embedded systems is still a major challenge. Although effective hardware/software co-design methods have been studied for decades, they are still not the standard in the development of complex embedded systems, including integrated AI systems. This paper discusses the open research challenges to realize such a holistic design space exploration for a hardware/software co-design for integrated AI systems. A novel approach for this holistic view of the hardware solutions is introduced, which is intended to evaluate accuracy, latency, as well as energy, and resource requirements. ... mehrIn Addition, this paper provides insights into the use case of an intelligent learning device for automated handwriting, which demonstrates an extremely resource-constrained environment.