In this article, we introduce a data-driven feedback linearizing controller for robotic manipulators with uncertain dynamics based on the augmented framework of time delay estimation (TDE) and a residue acceleration observer. While the time-delayed estimation approach in the inner-loop uses immediate past measurements of control effort and state to get an online estimate of the unknown manipulator dynamics, the residue acceleration observer in the outer-loop helps to nullify the estimation error on the fly, thus establishing a linear and decoupled dynamics. Once the linear dynamics is established, a PD controller helps to track the desired output trajectories. Simulation results for a two-link manipulator using the proposed controller, along with a delay-dependent convergence analysis, are shown to validate the proposition.

Ultrasound imaging is a commonly used modality for several diagnostic and therapeutic procedures. However, the diagnosis by ultrasound relies heavily on the quality of images assessed manually by sonographers, which diminishes the objectivity of the diagnosis and makes it operator-dependent. The supervised learning-based methods for automated quality assessment require manually annotated datasets, which are highly labour-intensive to acquire. These ultrasound images are low in quality and suffer from noisy annotations caused by inter-observer perceptual variations, which hampers learning efficiency. We propose an UnSupervised UltraSound image Quality assessment Network, US2QNet, that eliminates the burden and uncertainty of manual annotations. US2QNet uses the variational autoencoder embedded with the three modules, pre-processing, clustering and post-processing, to jointly enhance, extract, cluster and visualize the quality feature representation of ultrasound images. The pre-processing module uses filtering of images to point the network's attention towards salient quality features, rather than getting distracted by noise. Post-processing is proposed for visualizing the clusters of feature representations in 2D space. We validated the proposed framework for quality assessment of the urinary bladder ultrasound images. The proposed framework achieved 78% accuracy and superior performance to state-of-the-art clustering methods. The project page with source codes is available at https://sites.google.com/view/US2QNet.

Ultrasound is a vital imaging modality utilized for a variety of diagnostic and interventional procedures. However, an expert sonographer is required to make accurate maneuvers of the probe over the human body while making sense of the ultrasound images for diagnostic purposes. This procedure requires a substantial amount of training and up to a few years of experience. In this paper, we propose an autonomous robotic ultrasound system that uses Bayesian Optimization (BO) in combination with the domain expertise to predict and effectively scan the regions where diagnostic quality ultrasound images can be acquired. The quality map, which is a distribution of image quality in a scanning region, is estimated using Gaussian process in BO. This relies on a prior quality map modeled using expert's demonstration of the high-quality probing maneuvers. The ultrasound image quality feedback is provided to BO, which is estimated using a deep convolution neural network model. This model was previously trained on database of images labelled for diagnostic quality by expert radiologists. Experiments on three different urinary bladder phantoms validated that the proposed autonomous ultrasound system can acquire ultrasound images for diagnostic purposes with a probing position and force accuracy of 98.7% and 97.8%, respectively.

Modern robotic researches propose various machine vision methods for accomplishing robotic tasks. The recognition quality in these tasks is very important for successful performance. A large number of them use fiducial marker systems as a main element of algorithms. However, only a few researches are comparing standard marker systems. This paper is dedicated to the comparison of AprilTag and ArUco markers resistance to rotations in the presence of synthetic noise. Experiments were conducted in ROS/Gazebo virtual environment in order to provide a fair comparison of marker detection and recognition algorithms while eliminating external environment conditions that influence the algorithms' performance. The presented virtual environments allow collecting a significant amount of data by experiment process automation. Different levels of additive white Gaussian noise were applied to input sensory data in order to simulate the imperfection of real digital cameras. The main contribution of the paper is the systematic comparison of AprilTag and ArUco markers for rotation resistance in the presence of optical sensor noise.

Typical tasks of service robotics and special robotics fields, including Urban Search and Rescue, set a number of challenges for mobile robotics with automatic performance of various functions of mobile robots being a key task. To pursue automatic camera calibration processes using on-body markers in this paper we demonstrated results of validation experiments on mobile robot Servosila Engineer using fiducial markers, which we selected based on the results of virtual experiments with fiducial marker systems.

A new method is proposed for measurement of six dimensional pose of an industrial robot using a single image from a camera which is not pre-calibrated. Additionally during the pose determination, camera internal parameters are also obtained, which makes the method a suitable alternative for calibrating camera using a single image. Results from the two variants of the proposed approach are compared with Zhang's camera calibration algorithm and has been found to be better. Another utility of the proposed algorithm is to measure six dimensional pose of an industrial robot. Robot repeatability was also measured using the proposed camera calibration algorithm. The repeatability results were compared with the measurements using Artificial Reality toolkit ArUco which is a de-facto standard in the pose measurements using camera. The performance of the proposed method is better than that obtained from ArUco. Another area of application is identification of kinematic parameters. Using the circle point analysis method, identification of KUKA KR5 Arc robot was done using the proposed method.

We propose an approach for emptying of bin using a combination of Image and Range sensor. Offering a complete solution: calibration, segmentation and pose estimation, along with approachability analysis for the estimated pose. The work is novel in the sense that the objects to be picked are featureless and uniformly black in colour, hence existing approaches are not directly applicable. A key point involves optimal utilization of range data acquired from the laser scanner for 3-D segmentation using localized geometric information. This information guides segmentation of the image for better object pose estimation, used for pick-and-drop. We analytically assure the approachability of the object to avoid collision of the manipulator with the bin. Disturbance of objects caused during pick up has been modelled, which allows pickup of multiple pellets based on information from a single range scan. This eliminates the necessity of repeated scanning and data conditioning. The proposed method offers high object detection rate and pose estimation accuracy. The innovative techniques aimed at reducing the average pickup time makes it suitable for robust industrial operation.

Peg-in-tube assembly stands ahead of a more common benchmark task for industrial assembly, i.e., `pegin-hole'. The robot can easily be deceived to detect the actual hole while performing a `peg-in-tube' task as the tube has a surrounding pocket that cannot support the peg. The paper presents a thorough geometrical analysis of the `peg-in-tube' assembly process, and proposes a novel algorithm based on depth measurements of peg center to perform `peg-in-tube' task. The results are demonstrated on a KUKA KR5 Arc industrial robot with a chamferless cylindrical peg and a tube having a minimum clearance of 0.10 mm.

Active force control of an industrial robot with an end-effector force/torque sensor effectively handles compliant industrial tasks like assembly operations, surface finishing jobs, cooperative manipulation, etc. However, the robot still remains intrinsically unsafe for dynamically changing environment where the chances of the links coming in contact with the environment exists. This paper proposes a scheme for active force control at the end-effector using a six-component force/torque sensor through external force-control loop along with steady-state error compensator. In parallel, passive joint compliance was achieved by limiting the currents to joint motors based on an identified model of the robot under study. The proposed method was implemented on a KUKA KR5 ARC industrial robot and tested for passive compliance by colliding with another moving robot in its workspace. Active force control was tested to maintain a desired force on contact to demonstrate the effectiveness of the controller.

Cylinder, a common geometric entity has a discontinuity at the joining of cylindrical surface and circular-disks. Hence, collision detection between two cylinders in space is a difficult task and few have reported formulations to solve it. In this paper, a novel analytical methodology is proposed to detect collision or intersection between two cylinders. The configuration, i.e., position and orientation, between the cylinders was represented using the four Denavit-Hartenberg (DH) parameters plus two extra parameters. Dual Number Algebra was used to extract these six parameters. Tests involved in collision detection between the cylinders were between the lines and rectangles in a plane, thus considerably simplifying the problem of collision detection. As an illustration, an one-DOF arm modeled as a cylinder with cylindrical shaped obstacles were modeled and tested for their collisions. The results were validated with an analytical method available in the literature and a commercial software.

Constraint force minimization is essential to improve the dynamic performance of an industrial manipulator. An optimization method is proposed to minimize the constraint forces using the concept of dynamically equivalent system of point-masses. It is shown that for the six-DOF PUMA robot the constraint forces are substantially reduced.

A systematic methodology for the constraint forces analysis for designated joints is developed in a serial industrial manipulator. In the method, the system equations of motion are derived using the recursive Newton-Euler equations while constraint forces are determined non-recursively for designated joints. To validate the method, the six-DOF Stanford and PUMA robots are analysed. Accuracy of the results are compared using RIDIM (recursive inverse dynamics industrial manipulator), an in-house software. The same results are obtained from the proposed method and from the RIDIM

A matrix formulation of the constraint forces and moments, together referred here as the wrenches, for serial manipulators is presented in this paper. The methodology is based on the Newton-Euler dynamic equations of motions of the manipulator at hand. The formulation provides joint forces and moments that are necessary for designing the robot links, as well those required by the joint actuators. The latter information is required in inverse dynamics. The formulation is based on writing the kinematic constraint equations using the Decoupled Natural Orthogonal complement (DeNOC) matrices, introduced elsewhere. The kinematic information in the form of the DeNOC matrices of a serial manipulator is used for the determination of the joint torques and forces, which lead to the development of a uniform analytical tool for control and design.

The paper presents the dimensional design of a class of parallel manipulators, namely, Hexaslides. The design of hexaslides is formulated as a multiobjective optimization problem considering workspace and dexterity as dual objectives. As the relative emphasis on workspace and dexterity varies depending on the application, a set of Pareto-optimal solutions is found. The present analysis is a useful tool for designers to select suitable hexaslide parameters for a given application, particularly, in machine tool applications.

This paper presents kinematic analysis of a class of parallel manipulators, namely, Hexaslides, for machine tool applications. Hexaslides have constant-length legs. The inverse and direct kinematics solutions, to study the workspace properties of hexaslides are presented. Various kinematic performance indices, namely, workspace volume, workspace volume index, i.e. the ratio of workspace volume to machine size and global dexterity index of hexaslides having same foot print size are used to study different rail-arrangements. Secondly, effect of two important parameters related to configuration, i.e. gap between adjacent rails in case of hexaslide configuration and inclination of rails incase of slanted configuration, on the performance measures is presented.

A decomposition of the manipulator inertia matrix is essential, for example, in forward dynamics, where the joint accelerations are solved from the dynamical equations of motion. To do this, unlike a numerical algorithm, an analytical approach is suggested in this paper. The approach is based on the symbolic Gaussian elimination of the inertia matrix that reveal recursive relations among the elements of the resulting matrices. As a result, the decomposition can be done with the complexity of order n, O(n), where n being the degrees of freedom of the manipulator, as opposed to an O(n/sup 3/) scheme, required in the numerical approach. In turn, O(n) inverse and forward dynamics algorithms can be developed. As an illustration, an O(n) forward dynamics algorithm is presented.

An essential step in deriving kinematic models of free-flying space robots, consisting of a free-base and a manipulator mounted on it, is to write the total momenta of the system at hand. The momenta are, usually, expressed as the functions of the velocities of a preselected body that belongs to the robot, e.g., the free-base. In this paper, no preselection is recommended. On the contrary, the total momenta are expressed as the functions of the velocities of an arbitrary body of the space robot, namely, the primary body (PB). The identity of the PB, unlike the conventional approaches, need not be known at this stage. Therefore, the generalized expressions for the total momenta are obtained. The resulting expressions can explain the existing kinematic models and how they affect the efficiencies of the associated control algorithms. Based on the proposed approach, it is shown that if the end-effector motion is the only concern, as desired in kinematic control, it should be selected as the PB. This leads to the most efficient algorithms.

In this paper the UDU/sup T/ decomposition of the generalized inertia matrix of an n-link serial manipulator as presented in symbolic form, where U and D, respectively, are the upper triangular and diagonal matrices. To render the decomposition, the elementary upper triangular matrices, associated to a modified Gaussian elimination, are introduced, whereas each element of the inertia matrix is written as an expression, instead of finding it as a number with the aid of an algorithm. The resulting UDU/sup T/ decomposition shows recursive relations among the elements of the associated matrices. Thus, algorithms of order 'n' can be developed not only for the inverse but also for the forward dynamics. As an illustration, a forward dynamics algorithm is presented here.

A method based on the natural orthogonal complement is proposed for the development of dynamic models of space robots. As an illustration, the method is used to develop a computer simulation of a space manipulator that moves on a plane. The simulation results can be displayed using an animation program that has been developed in parallel to this research.

A systematic method for the kinematic and dynamic modeling of a two-degree-of-freedom (DOF) automatic guided vehicle (AGV) is presented. This type of methodology can be used to analyze, design, simulate, and control any kind of rolling robots. The concept of orthogonal complement is used to develop the dynamical equations of motion. The vehicle is analyzed for simulation purposes. Simulation results are reported. It is shown that, using the natural orthogonal complement of the matrix of velocity constraint equations, it is possible to derive systematically the Euler-Lagrange equations of motion of nonholonomic robotic mechanical systems. Moreover, the introduction of the orthogonal complement leads naturally to an efficient computational algorithm.>