Computing for Autonomy: Latency, Power, Resilience

Abstract

The articles in this special section focus on the development of autonomous systems. This theme issue on computing for autonomy presents responsive and reliable computing for assured autonomy. Autonomous systems have emerged as a key technology for a wide range of use cases. They can be used to inspect remote equipment, deliver packages, and transport people. Types of autonomous vehicles include those that operate on the ground, under the ocean, in the air, and in space. An autonomous system may include multiple vehicles. Vehicles may operate in formation or in swarms to perform a task. They may also cooperate to avoid collisions and traffic jams. Computing for autonomy clearly requires support for artificial intelligence and the large computational load that applications imply. These systems must also satisfy several types of nonfunctional requirements. Assured autonomy requires real-time responsiveness. Real-time operation is not only necessary for correct and efficient operation but also is a requirement for the safe operation of autonomous systems across many application domains. Safety is fundamental to the acceptance