Optical communication and guidance for maintenance & cargo drones at wind farms
Publieke samenvatting / Public summary
Aanleiding
The maintenance of offshore wind turbines and substations is expensive and logistically complex, requiring dedicated crew transfer vessels, motion-compensated walkways or even helicopter deployment of maintenance crew and materials. The use of autonomous drones to pre-deliver materials and maintenance equipment between ships and wind turbines/substations has the potential to make this process cheaper and safer, especially in harsh maritime environments. By using autonomous/robotic aerial vehicles, fewer or no human operators are needed. However, drone operations in offshore environments are highly challenging. Landing on ships in high sea states means landing on a moving platform that also changes attitude in unpredictable ways. Simultaneously, hoisting equipment to and from wind turbine nacelles is complicated by the spinning blades. Conditions are often windy and of reduced visibility, so a robust hovering / landing approach is needed that can deal with these disturbances. The drone needs real-time information on the wind turbine rotation status, wind conditions and ship motion. This information is crucial for a controlled descent, path planning and timing of the landing.
Doelstelling
The goal of this project is to use highly innovative active optical navigation and communication, digitization technologies and path planning and control strategies to support autonomous aerial vehicle operations (landing and hoisting) especially in harsh weather conditions (especially high winds, fog, haze, rain, etc.) in and around offshore wind farms. For a drone to land safely, it must know the relative position and orientation of the landing platform, its status, and gust conditions at the platform. Landing lights on ships and / or wind turbine nacelles can be used together with computer vision algorithms to extract this information. The lights should be designed such that the drone has a good view of the lights for navigation throughout the landing, even in adverse visibility conditions. Information will be transmitted via optical communication by flashing lights, transmitting a modulated signal. The drone can interpret this information embedded in the lighting and use it to make autonomous decisions.
Korte omschrijving
The activities are aimed at conceiving, developing and testing a light-based navigation and communication system for offshore drone operations near wind turbines and ships. 1. First, the requirements for the light system and the camera are defined, such that from the camera the relative pose can be determined from the lights and messages sent by flashing and or modulation of the lights can be received and interpreted. [Orga, TU Delft, DLR] 2. Next, the light system will be designed and built according to the requirements defined in the first work package, making use of existing lighting system designs as much as possible. [Orga] 3. In parallel to work package 2, in this work package a computer vision algorithm is developed for navigation and communication using the light system as defined in the first work package. This work package also includes the development of a guidance and control algorithm for landing. [TU Delft, DLR] 4. The goal of the last work package is to validate the integrated system. A camera-equipped drone will perform test approaches and landings, making use of the developed light system. [Aerogrid, TU Delft, DLR, Orga]
Resultaat
We propose a smart vertiport: a landing platform equipped with visual markers (lights) and smart drone that serve dual purposes: 1. The light sources will serve as an orientation beacon for the drones that come in for landing / hovering 2. The light sources will serve as a communication means for informing the drone on the local environmental and state information of the landing platform. The result of the project will be a design specification standard for smart vertiport systems for offshore drone operators. This standard will include specifications for the lights at the vertiport, their spatial layout, intensity, color, flashing and optical communication protocol, which will be made public. Computer vision algorithms will be developed and published as well, so that any drone with a camera can make use of the developed system. The optical communication and guidance system for maintenance & cargo drones at offshore wind farms will be validated and tested in the real world in representative conditions.
The maintenance of offshore wind turbines and substations is expensive and logistically complex, requiring dedicated crew transfer vessels, motion-compensated walkways or even helicopter deployment of maintenance crew and materials. The use of autonomous drones to pre-deliver materials and maintenance equipment between ships and wind turbines/substations has the potential to make this process cheaper and safer, especially in harsh maritime environments. By using autonomous/robotic aerial vehicles, fewer or no human operators are needed. However, drone operations in offshore environments are highly challenging. Landing on ships in high sea states means landing on a moving platform that also changes attitude in unpredictable ways. Simultaneously, hoisting equipment to and from wind turbine nacelles is complicated by the spinning blades. Conditions are often windy and of reduced visibility, so a robust hovering / landing approach is needed that can deal with these disturbances. The drone needs real-time information on the wind turbine rotation status, wind conditions and ship motion. This information is crucial for a controlled descent, path planning and timing of the landing.
Doelstelling
The goal of this project is to use highly innovative active optical navigation and communication, digitization technologies and path planning and control strategies to support autonomous aerial vehicle operations (landing and hoisting) especially in harsh weather conditions (especially high winds, fog, haze, rain, etc.) in and around offshore wind farms. For a drone to land safely, it must know the relative position and orientation of the landing platform, its status, and gust conditions at the platform. Landing lights on ships and / or wind turbine nacelles can be used together with computer vision algorithms to extract this information. The lights should be designed such that the drone has a good view of the lights for navigation throughout the landing, even in adverse visibility conditions. Information will be transmitted via optical communication by flashing lights, transmitting a modulated signal. The drone can interpret this information embedded in the lighting and use it to make autonomous decisions.
Korte omschrijving
The activities are aimed at conceiving, developing and testing a light-based navigation and communication system for offshore drone operations near wind turbines and ships. 1. First, the requirements for the light system and the camera are defined, such that from the camera the relative pose can be determined from the lights and messages sent by flashing and or modulation of the lights can be received and interpreted. [Orga, TU Delft, DLR] 2. Next, the light system will be designed and built according to the requirements defined in the first work package, making use of existing lighting system designs as much as possible. [Orga] 3. In parallel to work package 2, in this work package a computer vision algorithm is developed for navigation and communication using the light system as defined in the first work package. This work package also includes the development of a guidance and control algorithm for landing. [TU Delft, DLR] 4. The goal of the last work package is to validate the integrated system. A camera-equipped drone will perform test approaches and landings, making use of the developed light system. [Aerogrid, TU Delft, DLR, Orga]
Resultaat
We propose a smart vertiport: a landing platform equipped with visual markers (lights) and smart drone that serve dual purposes: 1. The light sources will serve as an orientation beacon for the drones that come in for landing / hovering 2. The light sources will serve as a communication means for informing the drone on the local environmental and state information of the landing platform. The result of the project will be a design specification standard for smart vertiport systems for offshore drone operators. This standard will include specifications for the lights at the vertiport, their spatial layout, intensity, color, flashing and optical communication protocol, which will be made public. Computer vision algorithms will be developed and published as well, so that any drone with a camera can make use of the developed system. The optical communication and guidance system for maintenance & cargo drones at offshore wind farms will be validated and tested in the real world in representative conditions.