February
Friday, 17th of February - 1pm (room Feynman)
Speaker: Giovanni Gozzini (visiting PhD student from Politecnico di Milano, Italy)
Title: Air-to-air landing of unmanned aerial vehicles
Abstract: It is well known that small electric Unmanned Aerial Vehicles (UAVs) suffer from low endurance problems. A possibility to extend the range of UAV missions could be to have a carrier drone with several lightweight multirotors (followers) aboard, which can take-off from and land on it.
In this presentation, the challenging problem of Air-to-Air Automatic Landing (AAAL) of UAVs is solved in two scenarios of increasing complexity. The first is a cooperative scenario in which the state of the carrier drone is perfectly known by the follower drone. A safe and fast landing is ensured by developing a strategy that combines quasi-time optimal feedback and a hybrid logic. In the second scenario, a non-cooperative approach is studied in which a vision-based state estimation technique onboard the follower is needed to predict the carrier motion and the previous hybrid QTO approach is combined with an observer-based adaptive law to ensure a safe and fast landing in a decentralized manner.
The proposed algorithms are validated through experimental activities involving landing a small quadcopter on a bigger octocopter used as a carrier.
Title: Air-to-air landing of unmanned aerial vehicles
Abstract: It is well known that small electric Unmanned Aerial Vehicles (UAVs) suffer from low endurance problems. A possibility to extend the range of UAV missions could be to have a carrier drone with several lightweight multirotors (followers) aboard, which can take-off from and land on it.
In this presentation, the challenging problem of Air-to-Air Automatic Landing (AAAL) of UAVs is solved in two scenarios of increasing complexity. The first is a cooperative scenario in which the state of the carrier drone is perfectly known by the follower drone. A safe and fast landing is ensured by developing a strategy that combines quasi-time optimal feedback and a hybrid logic. In the second scenario, a non-cooperative approach is studied in which a vision-based state estimation technique onboard the follower is needed to predict the carrier motion and the previous hybrid QTO approach is combined with an observer-based adaptive law to ensure a safe and fast landing in a decentralized manner.
The proposed algorithms are validated through experimental activities involving landing a small quadcopter on a bigger octocopter used as a carrier.
Speaker: Grégoire Le Goff (invited PostDoc from LAPLACE, Toulouse)
Title: Generalized Modelling and Control Allocation Methods in Power Electronics: the Modular Multilevel Converter (MMC) case
Abstract: Modular Multilevel Converters (MMCs) belong to the family of converters with a large number of switches. MMCs have the particularity of being comprised of sub-converters that interact in order to generate a given voltage. In this way, the MMC appears as an input redundant system and the control allocation methods appear as relevant solutions to address the control problem of such a system.
The work ranges from the modeling of the converter to the development of real-time optimization generic control allocation algorithms:
- Generalized and scalable modeling of the MMC,
- Development of the control allocation methods motivated by electrical engineering concerns,
- Scalable and generic control allocation of the MMC
Title: Generalized Modelling and Control Allocation Methods in Power Electronics: the Modular Multilevel Converter (MMC) case
Abstract: Modular Multilevel Converters (MMCs) belong to the family of converters with a large number of switches. MMCs have the particularity of being comprised of sub-converters that interact in order to generate a given voltage. In this way, the MMC appears as an input redundant system and the control allocation methods appear as relevant solutions to address the control problem of such a system.
The work ranges from the modeling of the converter to the development of real-time optimization generic control allocation algorithms:
- Generalized and scalable modeling of the MMC,
- Development of the control allocation methods motivated by electrical engineering concerns,
- Scalable and generic control allocation of the MMC