When it comes to using autonomous mobile robots (AMRs), drones or self-driving agricultural machines, it's currently difficult to develop and describe new complex tasks they need to perform. There's no clear or common language to define these tasks or the skills needed to complete them. If we had such a language, it would make these systems easier to understand, reuse, and adapt to new tasks.
Currently, the skills that robots and vehicles use are developed in a way that doesn’t allow their reuse; each skill is mostly built from scratch, and there are no frameworks to support skill composition. For example, an infrastructure inspection task for a drone needs multiple skills: navigation to a location, stabilizing in windy conditions, and actively moving to vantage positions to analyze infrastructure features in detail. However, there’s no best practice, design pattern or solution to easily reuse and combine individual skills with operational and safety interdependencies.
Project goal
The goal of this project is to enable companies to easily design, develop, deploy and operate a combination of (motion and task-specific) skills on different robot platforms with sufficient capabilities. To this purpose, a proof-of-concept framework containing best practices, design patterns and support tools for development and operation will be developed.
Would you also like to participate in this project?
AVAS_SBO is a Strategic Basic Research (SBO) project. We are looking for companies to join the User Group and work with us on the valorisation of the project.
Interested? Fill in the form below and we will contact you as soon as possible.
