Overview

About the project

To this end, the project will operationalize the concept of Purpose, drawn from the cognitive sciences, to increase the autonomy and domain independence of robots during autonomous learning and, at the same time, to lead them to acquire knowledge and skills that are actually relevant for operating in target real applications. In particular, the project will develop algorithms for the acquisition of purpose by the robot, ways to bias the perceptual, motivational and decision systems of the robots’ cognitive architectures towards purposes, and strategies for learning representations, skills and models that allow the execution of purpose-related deliberative and reactive decision processes. PILLAR-Robots will implement and validate demonstrators of purposeful lifelong open-ended autonomy using the resulting Purposeful Intrinsically Motivated Cognitive Architecture within three different application fields characterized by different types and levels of variability:

Objectives and aims

Objective 1: Operationalize the concept of Purpose to provide for a structured way to bias autonomy towards achieving the desires of the designer/user in a domain independent manner.Objectives and ambition

Objective 1.1 – Implement paths for the acquisition of purpose by the robot.
Objective 1.2 – Study and implement ways to bias the perceptual, motivational and decision systems of the robot cognitive architecture in the direction of purpose.
Objective 1.3 – Develop strategies for learning contextualized representations. skills and models within the cognitive architecture that allow executing purpose related deliberative/reactive decisions.

Objective 2: Create a reference purposeful Intrinsically Motivated Cognitive Architecture for lifelong open-ended autonomy.

Objective 2.1 – Hybridize MDB and GRAIL robot architectures from previous projects to create a base intrinsically motivated cognitive architecture.
Objective 2.2 – Study and implement ways to bias the perceptual, motivational and decision systems of the robot cognitive architecture in the direction of purpose.
Objective 2.3 – Develop strategies for learning contextualized representations. skills and models within the cognitive architecture that allow executing purpose related deliberative/reactive decisions.
Objective 2.4 – Provide APIs for the integration of simulators and industrially tested software components.
Objective 2.5 – Validate architecture in simulators and in reduced prototypes of the three applications considered in the project.

Objective 3: Implement and validate demonstrators of purposeful lifelong open-ended autonomy in real applications within three application domains characterized by different types and levels of variability.

Objective 3.1 – Validate Purposeful Intrinsically Motivated Cognitive Architecture in Agri-food scenario.
Objective 3.2 – Validate Purposeful Intrinsically Motivated Cognitive Architecture in Edutainment scenario.
Objective 3.3 – Validate Purposeful Intrinsically Motivated Cognitive Architecture in Industry/retail scenario.

Objective 4: Evaluate the possibilities and impacts of Purposeful Life Long Open-ended Autonomy in these application realms and in society in general (ethics, regulation, markets and a societal perspective).

Objective 4.1 – Study ethical implications of very high level of autonomy in robots.
Objective 4.2 – Propose remediation and suggest possible regulations.
Objective 4.3 – Evaluate possible markets and social impacts of these technologies.

Objective 5: Develop specific actions for fostering adoption and sustainability of the PILLAR-Robot framework across key stakeholders to maximise its impact by contributing to effective communication and training in AI – Data- Robotics.

Objective 5.1 – Build-up an IMOL/PIMOL community.
Objective 5.2 – Knowledge exchange with the ‘AI, Data and Robotics European Partnership’, other PPPs and platforms.
Objective 5.3 – Develop an exploitation and sustainability strategy to ensure the demonstration in TRL6-7 of the PILLAR exploitable assets.
Objective 5.4 – Define an Exploitation Plan
Objective 5.5 – Designing and delivering specific training program to the industrial community and other stakeholders.