Collaborative autonomous systems, such as smart warehouses, service robots, and autonomous vehicles, hold tremendous potential. To fully harness their capabilities, it is essential to process large volumes of data in real-time and make instantaneous decisions. Since these systems are safety-critical, mission-critical, and interact with people, the incorporation of trustworthy edge computing can be highly advantageous.
The goal of the COLA project is to facilitate the development of learning-based, multi-agent, real-time autonomous systems with humans in the loop by edge computing. The project specifically aims to focus on safety, performance, and social acceptability of collaborative autonomy applications.
In particular, the project links primarily to two TECoSA focus areas: Safety and Predictability.
Regarding safety, the aim is to develop novel techniques for analyzing safety and risk awareness in data-driven systems. This involves aligning provable safety measures with the perceived safety experienced by users in the context of human-in-the-loop systems. In terms of predictability, the project aims to investigate tractable, data-driven models of end-to-end system performance.
COLA naturally tightly collaborates with project SMEDE that provides the smart edge infrastructure, and particularly aims to feature its results primarily in the indoor testbed. It also collaborates with CART by development techniques relevant for connected, collaborative and automated road traffic as well.