Search

In this paper, an approach for fast and accurate segmentation of Deformable Linear Objects (DLOs) named FASTDLO is presented. A deep convolutional neural network is employed for background segmentation, generating a binary mask that isolates DLOs in the image. Thereafter, the obtained mask is processed with a skeletonization algorithm and the intersections between different DLOs are solved with a similarity-based network. Apart from the usual pixel-wise color-mapped image, FASTDLO also describes each DLO instance with a sequence of 2D coordinates, enabling the possibility of modeling the DLO instances with splines curves, for example. Synthetically generated data are exploited for the training of the data-driven methods, avoiding expensive collection and annotations of real data. FASTDLO is experimentally compared against both a DLO-specific approach and general-purpose deep learning instance segmentation models, achieving better overall performances and a processing rate higher than 20 FPS.

Deformable Linear Objects (DLOs) such as cables, wires, ropes, and elastic tubes are numerously present both in domestic and industrial environments. Unfortunately, robotic systems handling DLOs are rare and have limited capabilities due to the challenging nature of perceiving them. Hence, we propose a novel approach named RT-DLO for real-time instance segmentation of DLOs. First, the DLOs are semantically segmented from the background. Afterward, a novel method to separate the DLO instances is applied. It employs the generation of a graph representation of the scene given the semantic mask where the graph nodes are sampled from the DLOs center-lines whereas the graph edges are selected based on topological reasoning. RT-DLO is experimentally evaluated against both DLO-specific and general-purpose instance segmentation deep learning approaches, achieving overall better performances in terms of accuracy and inference time.

Within a robotic context, we merge the techniques of passivity-based control (PBC) and reinforcement learning (RL) with the goal of eliminating some of their reciprocal weaknesses, as well as inducing novel promising features in the resulting framework. We frame our contribution in a scenario where PBC is implemented by means of virtual energy tanks, a control technique developed to achieve closed-loop passivity for any arbitrary control input. Albeit the latter result is heavily used, we discuss why its practical application at its current stage remains rather limited, which makes contact with the highly debated claim that passivity-based techniques are associated to a loss of performance. The use of RL allows to learn a control policy which can be passivized using the energy tank architecture, combining the versatility of learning approaches and the system theoretic properties which can be inferred due to the energy tanks. Simulations show the validity of the approach, as well as novel interesting research directions in energy-aware robotics.

Nowadays robotic applications are widespread and most of the manipulation tasks are efficiently solved. However, Deformable-Objects (DOs) still represent a huge limitation for robots. The main difficulty in DOs manipulation is dealing with the shape and dynamics uncertainties, which prevents the use of model-based approaches (since they are excessively computationally complex) and makes sensory data difficult to interpret. This thesis reports the research activities aimed to address some applications in robotic manipulation and sensing of Deformable-Linear-Objects (DLOs), with particular focus to electric wires. In all the works, a significant effort was made in the study of an effective strategy for analyzing sensory signals with various machine learning algorithms. In the former part of the document, the main focus concerns the wire terminals, i.e. detection, grasping, and insertion. First, a pipeline that integrates vision and tactile sensing is developed, then further improvements are proposed for each module. A novel procedure is proposed to gather and label massive amounts of training images for object detection with minimal human intervention. Together with this strategy, we extend a generic object detector based on Convolutional-Neural-Networks for orientation prediction. The insertion task is also extended by developing a closed-loop control capable to guide the insertion of a longer and curved segment of wire through a hole, where the contact forces are estimated by means of a Recurrent-Neural-Network. In the latter part of the thesis, the interest shifts to the DLO shape. Robotic reshaping of a DLO is addressed by means of a sequence of pick-and-place primitives, while a decision making process driven by visual data learns the optimal grasping locations exploiting Deep Q-learning and finds the best releasing point. The success of the solution leverages on a reliable interpretation of the DLO shape. For this reason, further developments are made on the visual segmentation.

In this paper a new cable-driven haptic interface able to move in the six-dimensional space is presented. Both the design and the preliminary implementation of the interface are presented in this work. In particular, a 2-D setup is considered, with the aim of performing preliminary tests on the behaviour of this novel haptic system. Results regarding the capability of the system in rendering a virtual environment and in a robot teleoperation scenario with haptic force feedback are reported., https://youtu.be/EaS9xiHqGp4

This work considers an aerial robot system composed of an Unmanned Aerial Vehicle (UAV) and a rigid manipulator, to be employed in mobile manipulation tasks. The strategy adopted for accomplishing the aerial manipulation is a decomposition of the previous system in two decoupled subsystems: one concerning the center of mass of the aerial robot; and another concerning the manipulator's orientation. Two Lyapunov-based controllers are developed, using a back stepping procedure, for solving the trajectory tracking problems related to the two subsystems. In the controller design, three inputs are assumed available: a translational acceleration along a body direction of the UAV; an angular velocity vector of this body rotation; and, finally, a torque at the spherical, or revolute, joint connecting the UAV and the manipulator. The first two inputs are generated by the same controller in order to drive the center of mass on a desired trajectory; while a second controller drives, through the third input, the manipulator's orientation to track a desired orientation. Formal stability proofs are provided that guarantee asymptotic trajectory tracking. Finally, the proposed control strategy is experimentally tested and validated.

This paper explores effective algorithms for the solution of numerical nonlinear optimization problems in image restoration. The technology of modern acquisition techniques and devices most often returns data of increasing size, so we focus on the Scaled Gradient Projection algorithm, which is well suited to large-scale applications. We present its parallel implementations on different hardware, namely Tesla-series GPUs and cluster of last-generation multi-core CPUs. Algorithm and codes are tested against both 2D and 3D meaningful real-world problems.

We propose a new gradient projection algorithm that compares favorably with the fastest algorithms available to date for l1-constrained sparse recovery from noisy data, both in the compressed sensing and inverse problem frameworks. The method exploits a line-search along the feasible direction and an adaptive steplength selection based on recent strategies for the alternation of the well-known Barzilai-Borwein rules. The convergence of the proposed approach is discussed and a computational study on both well-conditioned and ill-conditioned problems is carried out for performance evaluations in comparison with five other algorithms proposed in the literature.

We propose a new gradient projection algorithm that compares favorably with the fastest algorithms available to date for {$\ell_1$}-constrained sparse recovery from noisy data, both in the compressed sensing and inverse problem frameworks. The method exploits a line-search along the feasible direction and an adaptive steplength selection based on recent strategies for the alternation of the well-known Barzilai-Borwein rules. The convergence of the proposed approach is discussed and a computational study on both well-conditioned and ill-conditioned problems is carried out for performance evaluations in comparison with five other algorithms proposed in the literature., SCOPUS: le.j, info:eu-repo/semantics/published