Home / Blog / Tech for Autonomous Systems – Autonomous Decision Taking Capabilities I

Tech for Autonomous Systems – Autonomous Decision Taking Capabilities I

Decision taking capabilities in the context of commercial flights. From relying completely on a set of rules to following orders from air traffic control, … up to full autonomy based on “see and avoid” decision taken by each individual aircraft. Image credit: JAPCC, Edition 20, 2015
  • Decision Making Technologies

 

A crucial aspect for autonomous systems, be it a drone, a manufacturing robot, a self driving car is the ability to take decisions in real time.

Notice that the decision making capability is actually a variety of degrees from a fully controlled system (from a third party taking full responsibility and living just the implementation to the system) from a full decision making by the autonomous system.

In case of a “human in the loop” we can see the human making all decision, to human becoming part of the system up to the human being one of the interacting systems, each one making its own decision (influencing and being influenced by other systems decisions). As an example till the end of last century the pilot made decision on deploying flaps and the aircraft would execute those decisions. Beginning with this century the aircraft aerodynamics has become so sophisticated that the tiny variations to the wings geometry can no longer be decided by the pilot and it is the aircraft that makes the decision based on inputs from the pilot (like altitude, speed, climbing rate…) and from sensors (like the pitot tube measuring airspeed). The pilot has become part of the system. In the future the geometry of the wing will be the result of several interacting systems, including some on the ground (determining the best flight attitude based on simulation of a variety of factors) and interacting with the plane and, possibly, with the pilot. There is no longer a hierarchy of systems but a variety of systems interacting with one another.

The role of humans in decision making will shift from a full control of the system, to the delegation of several implementation activity to the system to a partnership in the decision making process. Image Credit: Besser et al, Platform Autonomy, JAPCC, Edition 20, 2015

Decision making technologies rely on perception and awareness, recognition, learning, planning, knowledge representation and reasoning.

The decision making process combines pre-loaded data (a-priori knowledge), sensors data processed to create awareness, interaction with other autonomous systems, including humans, and infers goals and plans of other autonomous systems in its environment to create a set of executable actions. Actions are taken on the bases of an overall goal that needs to be achieved by the autonomous systems. The validity of that goal is not within the autonomous system to dispute, however, the means through which that goal can be pursued is within the autonomous decision power and this brings to the fore ethical issues, that in turns have an impact on the decision making process and on the technologies used.

Several products offering an integrated set of technologies for decision making exist on the market and more will become available in the coming years as the number of autonomous systems and application areas will keep growing.

DARPA is funding several research aiming at decision making in autonomous systems in a variety of situations and context:

  • ONR LOCUST (Low Cost UAV Swarming Technologies) aiming at enabling decision making in drones, normally controlled by an operator but able to act on their own, by collaborating with one another if connection to the operator is lost.
  • ONR CARACaS (Control Architecture for Robotic Agent Command and Sensing) aiming at providing decision capability to autonomous boats for patrolling harbors and escorting ships.
  • DARPA ALIAS (Aircrew Labor in Copckpit Automation System) aiming at reducing the work load of pilot in military and commercial aircraft providing autonomous decision making capability.

About Roberto Saracco

Roberto Saracco fell in love with technology and its implications long time ago. His background is in math and computer science. He’s currently Head of the Industrial Doctoral School of EIT Digital, co-chair of the Symbiotic Autonomous Systems Initiative of IEEE-FDC. Until Aprile 2017 he led the EIT Digital Italian Node. Previously, up to December 2011 he was the Director of the Telecom Italia Future Centre in Venice, looking at the interplay of technology evolution, economics and society. At the turn of the century he led a World Bank-Infodev project to stimulate entrepreneurship in Latin America. He is a senior member of IEEE where he leads the Industry Advisory Board within the Future Directions Committee. He teaches a Master course on Technology Forecasting and Market impact at the University of Trento.
He has published over 100 papers in journals and magazines and 14 books.
He writes a daily blog,  http://sites.ieee.org/futuredirections/category/blog/, with commentary on innovation in various technology and market areas.

Leave a Reply