SubsidieMeestersAutonomous Robots with Common Sense
This project aims to develop an 'Artificial Physical Awareness' autopilot system for autonomous robots, enabling them to operate safely and effectively despite failures by understanding their limitations.
Projectdetails
Introduction
Autonomous robots such as autonomous vehicles, cars, and drones have the potential to revolutionize the way we work and live. Unfortunately, current autonomous robots do not have ‘common sense’ and may enter a catastrophic condition called loss-of-control after failures. The ultimate goal of this research is to enable a new generation of autonomous robots that are aware of their physical capabilities and limitations, allowing them to act with common sense after failures.
Proposed Paradigm
To achieve this, I propose a new paradigm in autonomous robot control: ‘Artificial Physical Awareness’ (APA). APA requires accurate real-time knowledge of the time-varying stochastic safe envelope, which is a subset of the state-space inside which safe operations of the autonomous robot can be guaranteed.
Characteristics of the Safe Envelope
The safe envelope is stochastic and time-varying; it contains uncertainties and will shrink after failures, reflecting the reduced post-failure performance of the autonomous robot.
Scientific Challenge
Obtaining and utilizing the time-varying stochastic safe envelope in real-time represents a currently unsolved scientific challenge for the following reasons:
- The safe envelope cannot be measured directly.
- Current safe envelope computation methods are real-time intractable and/or do not take into account uncertainties.
- No control methodology exists that allows for time-varying safe-envelope informed balancing of safety and performance.
Multidisciplinary Approach
This multidisciplinary research combines new insights in:
- Time-varying stochastic state reachability analysis
- Tipping-point forecasting
- Bio-inspired envelope sensing and recovery
- Nonlinear fault-tolerant control
to develop the new APA-autopilot system, which is the main output of this research.
Potential Impact
This research project has the potential to lead to a revolution in autonomous robot design and operations by providing transparent safety and performance bounds, even after failures.
Financiële details & Tijdlijn
Financiële details
| Subsidiebedrag | € 1.996.040 |
| Totale projectbegroting | € 1.996.040 |
Tijdlijn
| Startdatum | 1-7-2024 |
| Einddatum | 30-6-2029 |
| Subsidiejaar | 2024 |
Partners & Locaties
Projectpartners
- TECHNISCHE UNIVERSITEIT DELFTpenvoerder
Land(en)
Vergelijkbare projecten binnen European Research Council
| Project | Regeling | Bedrag | Jaar | Actie |
|---|---|---|---|---|
The Artificial Motion FactoryARTIFACT aims to revolutionize robot autonomy by developing a modular AI control architecture that enables advanced decision-making and interaction in dynamic environments through learning and perception. | ERC Starting... | € 1.499.955 | 2025 | Details |
Breaching the boundaries of safety and intelligence in autonomous systems with risk-based rationalityThis project aims to develop a comprehensive risk-based autonomy framework for autonomous systems, enhancing safety and decision-making in marine environments through advanced modeling and human supervision. | ERC Advanced... | € 2.499.773 | 2025 | Details |
Intuitive interaction for robots among humansThe INTERACT project aims to enable mobile robots to safely and intuitively interact with humans in complex environments through innovative motion planning and machine learning techniques. | ERC Starting... | € 1.499.999 | 2022 | Details |
Automated Synthesis of Certifiable Control Software for Autonomous VehiclesCertiCar aims to develop a reliable, formally correct advanced collision avoidance system to enhance safety and reduce testing time for autonomous vehicle control software. | ERC Proof of... | € 150.000 | 2024 | Details |
Deep Bayesian Reinforcement Learning -- Unifying Perception, Planning, and ControlDevelop an algorithmic framework using deep learning and Bayesian reinforcement learning to enhance robotic manipulation in unstructured environments by effectively managing uncertainty. | ERC Starting... | € 1.500.000 | 2022 | Details |
The Artificial Motion Factory
ARTIFACT aims to revolutionize robot autonomy by developing a modular AI control architecture that enables advanced decision-making and interaction in dynamic environments through learning and perception.
Breaching the boundaries of safety and intelligence in autonomous systems with risk-based rationality
This project aims to develop a comprehensive risk-based autonomy framework for autonomous systems, enhancing safety and decision-making in marine environments through advanced modeling and human supervision.
Intuitive interaction for robots among humans
The INTERACT project aims to enable mobile robots to safely and intuitively interact with humans in complex environments through innovative motion planning and machine learning techniques.
Automated Synthesis of Certifiable Control Software for Autonomous Vehicles
CertiCar aims to develop a reliable, formally correct advanced collision avoidance system to enhance safety and reduce testing time for autonomous vehicle control software.
Deep Bayesian Reinforcement Learning -- Unifying Perception, Planning, and Control
Develop an algorithmic framework using deep learning and Bayesian reinforcement learning to enhance robotic manipulation in unstructured environments by effectively managing uncertainty.
Vergelijkbare projecten uit andere regelingen
| Project | Regeling | Bedrag | Jaar | Actie |
|---|---|---|---|---|
Symbolic logic framework for situational awareness in mixed autonomySymAware aims to develop a comprehensive framework for situational awareness in multi-agent systems, enhancing collaboration and safety between autonomous agents and humans through advanced reasoning and risk assessment. | EIC Pathfinder | € 3.980.291 | 2022 | Details |
Counterfactual Assessment and Valuation for Awareness ArchitectureThe CAVAA project aims to develop a computational architecture for awareness in biological and technological systems, enhancing user experience through explainability and adaptability in various applications. | EIC Pathfinder | € 3.132.460 | 2022 | Details |
Autonoom navigerenRoboat en Praxis Automation ontwikkelen een autonoom navigatiesysteem voor schepen dat gebruikmaakt van AI en sensortechnologie om de veiligheid en efficiëntie in de maritieme sector te verbeteren. | Mkb-innovati... | € 249.550 | 2023 | Details |
Fail-operational safety – making autonomous vehicles a realityChassis Autonomy aims to finalize a fail-operational steer-by-wire system for fully autonomous vehicles, enabling driverless technology and targeting market launch in two years. | EIC Accelerator | € 2.497.305 | 2023 | Details |
Perception of Collaborative RobotsHet project onderzoekt de haalbaarheid van technieken zoals voice control en machine vision om collaboratieve robots beter omgevingsbewust te maken voor gebruik in high-mix low-volume productie. | Mkb-innovati... | € 20.000 | 2020 | Details |
Symbolic logic framework for situational awareness in mixed autonomy
SymAware aims to develop a comprehensive framework for situational awareness in multi-agent systems, enhancing collaboration and safety between autonomous agents and humans through advanced reasoning and risk assessment.
Counterfactual Assessment and Valuation for Awareness Architecture
The CAVAA project aims to develop a computational architecture for awareness in biological and technological systems, enhancing user experience through explainability and adaptability in various applications.
Autonoom navigeren
Roboat en Praxis Automation ontwikkelen een autonoom navigatiesysteem voor schepen dat gebruikmaakt van AI en sensortechnologie om de veiligheid en efficiëntie in de maritieme sector te verbeteren.
Fail-operational safety – making autonomous vehicles a reality
Chassis Autonomy aims to finalize a fail-operational steer-by-wire system for fully autonomous vehicles, enabling driverless technology and targeting market launch in two years.
Perception of Collaborative Robots
Het project onderzoekt de haalbaarheid van technieken zoals voice control en machine vision om collaboratieve robots beter omgevingsbewust te maken voor gebruik in high-mix low-volume productie.