GRVC Group - University of Seville

https://grvc.us.es/


Power Line Inspection Tasks with Multi-Aerial Robot Systems 

The main objective of the proposed project is to compute feasible and constrained trajectories for a fleet of multi-rotor aerial vehicles (i.e., under-actuated vehicles) that have been assigned an inspection task together by a High-Level Planner. The inspection tasks are indicated with the their spatial coordinates in the 3D space (x, y, and z) and the heading angle (\psi), necessary to orientate the themselves in the environment, along with the action to take (e.g., record a video or take a picture). The multi-rotors operate in a known environment, represented by a map (i.e., an occupancy grid map) that also includes the position of obstacles and the power tower, and are located using RTK GPS sensors. In the multi-rotor hardware setup, cameras are mounted in an eye-in-hand configuration, i.e., rigidly attached to the body frame.

An offline has been recently proposed to cope with the task [1]. The set up optimization problem computes trajectories that avoid obstacles and maintain a safe distance between drones while also accounting for their velocity and acceleration constraints. The algorithm divides the inspection tasks among the available vehicles to generate optimal strategies that satisfy the mission specifications. The High-Level Planner monitors the plan execution, integrating information perceived by the drone (i.e., their battery level, localization, and target state estimation), in order to re-plan online the mission if the initial plan is no longer feasible. In addition, this information is used to decide when a drone has failed and needs to be landed urgently. For instance, to replace a vehicle that may run out of battery before finishing its current task. A video with the preliminary numerical simulations in Gazebo and real flight tests are shown below. This task is part of the work carried out in the context of the European project AERIAL-CORE.

[1] G. Silano, T. Baca, R. Penicka, D. Liuzza and M. Saska, "Power Line Inspection Tasks With Multi-Aerial Robot Systems Via Signal Temporal Logic Specifications," in IEEE Robotics and Automation Letters, vol. 6, no. 2, pp. 4169-4176, April 2021, doi: 10.1109/LRA.2021.3068114.

 


Surveilling Tasks for Monitoring and Asssisting Human Operators in Safety Critical Scenarios

A fleet of multi-rotor aerial-vehicles (i.e., under-actuated vehicles) have been assigned a surveilling task together by a High-Level Planner. The vehicles are required to monitor a human being working on an electrical tower ensuring its safety, i.e., images and videos are provided to the ground supervision team who assist the operator working on the tower. The vehicles operate in a known environment, represented by a map (i.e., an occupancy grid map) that also includes the position of obstacles and the power tower. Both the drones and the human worker are located using RTK GPS sensors. In the vehicle hardware setup, cameras are mounted in an eye-in-hand configuration, i.e., rigidly attached to the body frame. An online running planner computes feasible trajectories for the drones working in a leader-follower scheme. Obstacle avoidance capabilities guarantee compliance with safety requirements between the multi-rotors and the human worker. The High-Level Planner monitors the plan execution, integrating information perceived by the drones (i.e., their battery level, localization, and target state estimation), in order to replan online the mission if the initial plan is no longer feasible. In addition, this information is used to decide when a drone has failed and needs to be landed urgently. For instance, to replace a vehicle that may run out of battery before finishing its current task. Moreover, the human worker can use gestures to interact with the multi-rotors by requesting for high-level actions, such as scaling the formation or change the camera viewing angles. In this case, the High-Level Planner takes care of recomputing the mission plan. A video with the preliminary numerical simulations in Gazebo and real flight tests is shown below. This task is part of the work carried out in the context of the European project AERIAL-CORE.

[2] V. Krátký, A. Alcántara, J. Capitán, P. Štěpán, M. Saska and A. Ollero, "Autonomous Aerial Filming With Distributed Lighting by a Team of Unmanned Aerial Vehicles," in IEEE Robotics and Automation Letters, vol. 6, no. 4, pp. 7580-7587, Oct. 2021, doi: 10.1109/LRA.2021.3098811.


Publications

  1. Vít Krátký, Alfonso Alcántara, Jesús Capitán, Petr Štěpán, Martin Saska and Aníbal Ollero. Autonomous Aerial Filming with Distributed Lighting by a Team of Unmanned Aerial Vehicles. IEEE Robotics and Automation Letters 6(4):7580-7587, October 2021. PDF, DOI BibTeX

    @article{kratky2021aerialfilming,
    	author = "{Krátký}, Vít and {Alcántara}, Alfonso and {Capitán}, Jesús and {Štěpán}, Petr and {Saska}, Martin and {Ollero}, Aníbal",
    	title = "{Autonomous Aerial Filming with Distributed Lighting by a Team of Unmanned Aerial Vehicles}",
    	journal = "IEEE Robotics and Automation Letters",
    	year = 2021,
    	month = "October",
    	pdf = "data/papers/aerialCinematographyRal.pdf",
    	doi = "10.1109/LRA.2021.3098811",
    	pages = "7580-7587",
    	volume = 6,
    	number = 4
    }
    
  2. Giuseppe Silano, Jan Bednar, Tiago Nascimento, Jesus Capitan, Martin Saska and Anibal Ollero. A Multi-Layer Software Architecture for Aerial Cognitive Multi-Robot Systems in Power Line Inspection Tasks. In 2021 International Conference on Unmanned Aircraft Systems (ICUAS). June 2021, 1624–1629. PDF, DOI BibTeX

    @inproceedings{Silano2021ICUAS-I,
    	author = "{Silano}, Giuseppe and {Bednar}, Jan and {Nascimento}, Tiago and {Capitan}, Jesus and {Saska}, Martin and {Ollero}, Anibal",
    	booktitle = "2021 International Conference on Unmanned Aircraft Systems (ICUAS)",
    	title = "{A Multi-Layer Software Architecture for Aerial Cognitive Multi-Robot Systems in Power Line Inspection Tasks}",
    	doi = "10.1109/ICUAS51884.2021.9476813",
    	year = 2021,
    	month = "June",
    	pdf = "data/papers/ICUAS21_Silano_I.pdf",
    	pages = "1624--1629",
    	organization = "IEEE"
    }