REXASI-PRO
The project
About the project
About the project
About REXASI-PRO
To design a novel framework in which safety, security, ethics, and explainability are entangled to develop a Trustworthy Artificial Swarm Intelligence solution.
The framework will make a trustworthy collaboration among a swarm formed by autonomous wheelchairs and flying robots to allow a seamless door-to-door experience for people with reduced mobility.
Expected Impact
The impact of REXASI-PRO is to deliver a new Trustworthy AI framework. This framework will enable the collaboration among a swarm formed by autonomous wheelchairs, and flying robots to enable a seamless door-to-door experience for people with reduced mobility.
This goal will result in benefits for these people, their families, caregivers, scientific community, industry, and environment, creating a scientific, economic, technological and social benefit.
What we Do
What we Do
Work plan and resources #1
Considering the project overall goals, to guarantee the objectives achievement and to manage the project efficiently and effectively, a coherent work-plan, over 36 months, has been devised. REXASI-PRO work-plan is organized through 8 work-packages (WPs). WP2 will set the reference context for the project and will collect all the requirements (technical, ethical, performance, integration,) to develop the REXASI-PRO framework. In addition, it will set the procedure to assess GDPR compliance throughout project execution.
- in WP3 novel trustable algorithmic solutions for social navigation will be designed, in addition the project will define novel vocal-based interaction to maximize user’s satisfaction.
- WP4 will explore novel formal approaches to guarantee that the algorithm design by WP3 are robust, safe and secure;
- WP5 will increase the reliability of
the algorithms by adding explainability to the Verification and Validation methodologies, while; - WP6 will keep focused on making the REXASI framework greener. In addition, WP6 will showcase how novel topological method can be used to:
– Reduce ML power consumption by optimizing the training dataset
– The ML solutions do not contain any biases.
How we do IT
Work plan and resources #2
The REXASI-PRO Framework obtained by this modular architecture will be deployed in WP7 into the autonomous wheelchair use-case, leading finally to the testing, validation, and demonstration of the trustable social navigation in two different real-life scenarios:
- Navigating an environment equipped by redundant infrastructure that can provide additional data to the wheelchair;
- Navigating an environment without the infrastructure. WP8, Impact creation, exploitation and dissemination, and WP1, project management, will run in parallel throughout the other project WPs.
Use cases
Autonomous wheelchair
AI for Autonomous Wheelchair in different scenarios:
1) Safety Assistant;
2) Driving Assistant;
3) Route Assistant; and
4) Social Navigation.
To adopt this technology in real-life scenario, new trustable social navigation approaches are required. Among the challenges, the wheelchair shall adapt to the dynamics of the environment (avoid collisions, avoid unexpected displacements), shall maximize driver well-being, guarantee indoor localization and target identification.
Autonomous flying drones
Flying robots capable of flying autonomously and replacing the need for human intervention within indoor/underground environments.
The robots can fly in absence of light, Global Navigation Satellite System (GNSS) or other kind of wireless radio communication system. This technology will be used to generate the 3D map of the environment without any previous knowledge of its dimension or layout. The generated map will be later used to generate a siulation environement and by the wheelchair to navigate.
Drone & Wheelchair
Mixed collaborative indoor environment where the 2 swarm communicate with each other in emergency cases.