In this paper, we present selected benchmark aerial manipulation tasks using an aerial vehicle endowed with multi-degree of freedom manipulators. The proposed tasks analyze environmental coupling and are broken into three general categories: momentary, loose, and strong coupling. A classical control structure is derived, tuned, and verified through experiments, conducted for benchmarking purposes to include pick-and-place, insertion, and valve-turning tasks. Although other nonlinear controllers may prove more effective, the classical control approach has been selected in order to analyze contact stability and provide benchmark results for future reference. An analysis of system stability is conducted and implemented into the controller. A vision-based high-level controller fuses motion tracking data in order to provide control of both the aircraft and the manipulators, allowing the system to become coupled to the environment and perform the required operation. We present recent results validating our framework using the proposed aircraft-arm system.

This paper presents a LiDAR-equipped unmanned aerial vehicle (UAV) performing semi-autonomous wind-turbine blade inspection which includes traversing to the blade tip and back, while keeping constant relative distance and heading to the blade plane. Plane detection is performed applying the RANSAC method on a subset of the gathered pointcloud. Utilizing the relative distance to the inferred plane as well as its normal vector, the UAV is able to maintain a constant distance and heading towards the plane while moving in parallel with it. The proposed procedure performs successful wind-turbine blade inspections with minimal operator involvement. Inspection results include high-resolution blade images as well as a 3D model of the wind-turbine structure. Finally, to show the feasibility of this approach, simulations are performed on a wind-turbine model and experimental results are presented for an outdoor single-blade inspection scenario both on a mock-up setup and a full-scale wind-turbine blade. It is worth noting that the system’s adequacy has been fully validated in real conditions on an operational wind farm.

A common approach in the field of tactile robotics is the development of a new perception algorithm for each new application of existing hardware solutions. In this letter, we present a method of dimensionality reduction of an optical-based tactile sensor image output using a convolutional neural network encoder structure. Instead of using various complex perception algorithms, and/or manually choosing task-specific data features, this unsupervised feature extraction method allows simultaneous online deployment of multiple simple perception algorithms on a common set of black-box features. The method is validated on a set of benchmarking use cases. Contact object shape, edge position, orientation, and indentation depth are estimated using shallow neural networks and machine learning models. Furthermore, a contact force estimator is trained, affirming that the extracted features contain sufficient information on both spatial and mechanical characteristics of the manipulated object.

This letter presents a collaborative human-robot framework for delicate sanding of complex shape surfaces. Delicate sanding is performed using a standard industrial manipulator, equipped with the force/torque sensor and specially designed compliant control algorithm. Together with the compliant control, we discuss trajectory planning and safety problem of such an approach. The experience of the human workers is exploited through the intuitive framework and applied to plan the trajectories for the robot. The flexibility and the reliability of the proposed framework is tested in real working conditions in the factory.

The power grid is an essential infrastructure in any country, comprising thousands of kilometers of power lines that require periodic inspection and maintenance, carried out nowadays by human operators in risky conditions. To increase safety and reduce time and cost with respect to conventional solutions involving manned helicopters and heavy vehicles, the AERIAL-CORE project proposes the development of aerial robots capable of performing aerial manipulation operations to assist human operators in power lines inspection and maintenance, allowing the installation of devices, such as bird flight diverters or electrical spacers, and the fast delivery and retrieval of tools. This manuscript describes the goals and functionalities to be developed for safe local aerial manipulation, presenting the preliminary designs and experimental results obtained in the first year of the project.