The paper deals with non-standard motion tasks specified for a two-wheeled nonholonomic robot. It is assumed that wheels cannot fully rotate which reduces a set of feasible movements significantly. In spite of these constraints, it is expected that position of the robot can be changed without violating nonholonomic constraints. Such a possibility comes from the small time local controllability (STLC) of the kinematics described on four-dimensional configuration manifold. In order to solve these specific tasks a feedback taking advantage of the transverse function approach is designed. Consequently, the system can be virtually released from non-holonomic constraints. The transverse function also defines a virtual geometry constraint which makes it possible to limit wheels rotation.Properties of the designed controller are illustrated by results of numerical simulations in various motion task scenarios.

Sequence stabilization of medical images is an important aspect of diagnosis, therapy, joint movement kinematic analysis, and cancer detection. Typically, when image frames are recorded, the body is not rigidly fixed as a result of e.g. respiration, thus the position of its segments may vary. Simple image analysis methods (e.g. gradient based, scale-space based) tend to have problems with discerning the key-points in this specific task, due to large diversity of bone structure and highly visible soft tissue. In this paper, we propose a specialized algorithm for stabilization of femur in a sequence of single plane fluoroscopic images. The method estimates the positions of several easily-detectable femur key-points using gradient-based image analysis methods. For other key-points, which are located in the regions of bone with saliency prohibiting effective detection, we use feedforward Convolutional Neural Network as a position estimator. All the key-point positions are used in a stabilization process performed with the ICP (Iterative Closest Point) algorithm. The overall stabilization accuracy is evaluated for two uncorrelated X-ray image sequences, where manual stabilization (i.e., the results for image alignment performed by a human operator without access to key-points) constitutes the ground truth.

Underground coal mining is a branch of an industry which safety of operation is very dependent on the natural hazards.A proper seismic event prediction is a significant aspect of building classification models from the real data, which can affect the coal mining safety increase.In this paper four models, built in a well known data mining environments, are presented.The obtained models, depending on a given implementation of popular methods, occurred comparable to the best results from the competition.

In this paper a unified motion control and planning algorithm dedicated for the waypoint following task realized by a skid-steering is presented. In order to reduce excessive slip effects between robot wheels and ground it is assumed that the vehicle moves with bounded velocities and accelerations. The motion controller has been designed using formal methods to ensure asymptotically stable tracking of feasible reference trajectories. To account for practical motion tasks the trajectory tracking algorithm is complemented by a motion planner, which utilizes the differential flatness property of unicycle-like kinematics. During the motion planning stage an auxiliary trajectory connecting points in the configuration space and satisfying assumed phase constraints is generated. The resulting motion execution system has been implemented on a laboratory-scale skid-steering mobile robot, which served as platform for experimental validation of presented algorithms.

The paper considers application of a controller using transverse functions to solving the regulation and trajectory tracking problem for a multi-body vehicle which consists of a unicycle-like tractor with three trailers. In order to facilitate the controller design, the kinematics of the system is partially defined on the Lie group. The control solution is based on configuration and input transformation of an open-loop error dynamics to the invariant system with a drift. The properties of the closed-loop control system are examined based on the results of numerical simulations considering the selected motion tasks.

The paper considers the convergence problem for a two input affine nonholonomic driftless system with three-dimensional state. The problem is solved using time-invariant static-state feedback and polar transformation which is singular at the origin. The given solution, in general, is local, and the feasible domain is dependent on the properties of the control system. Theoretical considerations are illustrated for the particular system and the results of numerical simulations are presented.

The paper presents a comparison of selected smooth and time-differentiable control laws used for stabilization of nonholonomic two-wheeled mobile robot. The significant part of the research is concentrated on application of these algorithms in resolving parallel parking problem. Some remarks concerning tuning of the controllers and their robustness to the delay in control loop are given

The application of a novel branch of computational cybernetics is extended to the adaptive control of very inaccurately and partly modeled electromechanical systems also having unmodeled non-linear dynamic coupling with their environment. The method uses "uniform structures" for modeling, and "uniform procedures" for "repetitive" or cumulative learning like in the case of the "traditional" soft computing approaches. But it considerably reduces the number of free parameters tuning with simple, lucid, and explicit algebraic operations of limited number of steps. It is demonstrated via simulation that the control can operate even if strongly nonlinear terms (Coulomb friction and sticking combined with the elastic and viscous terms of the unmodeled external interaction in the Stribeck model), and elastically deformable joints are present. As a paradigm a mechanically 3 DOF SCARA arm actuated by voltage-controlled DC motors is considered. The system has 9 degrees of freedom (6 mechanical and 3 electrical). For the controller only three mechanical degrees can efficiently compensate the effect of the unmodeled degrees of freedom, the external interaction, and the inaccuracy of the modeled part of the robot.

A trajectory tracking control problem for a nonholonomic mobile robot by making use of a kinematic oscillator has been solved. Firstly, a time varying oscillator is examined to control the nonholonomic mobile robot based only its kinematics. Secondly, a backstepping procedure is proposed to include robot dynamics and the servo loop. It is shown that overall multilevel controller is asymptotically globally stable to a small error different from zero. A wide range of simulation results are presented which illustrate the behaviour of the controller with respect to tuning its parameters. Some preliminary experiments are reported too.

This paper deals with the properties of a dynamical systems expressed in terms of so called the eigen-factor quasi-coordinate velocities. Using these variables we can diagonalize the mass matrix of a manipulator which implies that at each fixed time instant each joint equation is decoupled from all of the other joint equations. It is shown that the structure of dynamic equations of motion in terms of the eigen-factor quasi-coordinate velocities enables different insights into the manipulator behavior as compared to classical equations. We point out differences between the two formulations.

This paper presents a comparison of PD controls in joints space for serial manipulators whose dynamics is expressed in terms of two different kinds of quasi-velocities. Robot dynamic algorithms in terms of so called normalized and unnormalized quasi-velocities are recursive in nature and consists of two recursions: one starts from a base of the manipulator towards its tip, and the other in an opposite direction. Both recursions are described by using a vector-matrix notation. An instantaneous eigenstructure quasi-velocity formulation, as a result of numerical mass matrix factorization, is introduced. The two PD controls were tested on the model of a manipulator with two degrees of freedom. The standard control algorithms known in robotic literature are considered for a case when the robot dynamics is formulated in terms of quasi-velocities.

This paper presents some remarks on PD control in joint space for serial manipulators whose dynamics is expressed in terms of two different kind of quasi-velocities. Robot dynamics algorithms in terms of so called normalized quasi- velocities are recursive in nature and consists of two recursions: one starts from a base of the manipulator towards its tip and the second in opposite direction. Both recursions are described by making use of vector-matrix notation. In the second work authors introduce an instantaneous eigenstructure quasi-velocity formulation as a result of numerical mass matrix factorization. Both PD controls were tested on a model of manipulator with two degrees of freedom. In this paper we adopt standard PD control known in robotic literature to a case when robot dynamics is formulated in terms of quasi-velocities.

Traditional fuzzy controllers or artificial neural networks are constrained by standardized formal procedures and structural restrictions. While in general it is guaranteed that these approaches or their integration can solve a quite wide class of problems, their "orthodox" application may result in too complicated control not exploiting the peculiarities of the given task under consideration. The aim of the paper is to demonstrate that unconventional integration of simple elements such as classic PID/ST, regression analysis, saturated sigmoid transition functions, fuzzy sets and uniform structures obtained from the Lagrangian classical mechanics instead of the connection structure of a feedforward artificial neural networks can result in a simple and efficient adaptive control for robots involved in unknown environmental dynamic interaction. Due to their simplicity and reduced number of parameters real time tuning can be carried out in these structures. Most of the parameters are independent of the particular problem to be solved and neither "scaling problems" nor "network paralysis" occur during the learning phase. It is concluded that the different components of the control can successfully co-operate in finding the "proper" system model even in the case of very rough initial model estimation and external interaction.

This paper presents a comparison of controls for manipulators whose dynamics is expressed in terms of quasi-velocities. Robot dynamic algorithms in terms of quasi-velocities are recursive in nature and consists of two recursions: one starts from a base of the manipulator towards its tip and the second in opposite direction. Both recursions are described by using vector-matrix notation. Quasi-velocities are a result of numerical mass matrix factorization. They were tested on a model of manipulator with two degrees of freedom. In this paper we adopt standard control algorithms known in robotic literature to a case when robot dynamics is formulated in terms of quasi-velocities.

In this paper, a simple one degree of freedom robot, designed for both research and teaching purposes, is described. The experimental set-up consists of a DC-motor, rate generator, harmonic drive, control system and load. A system has an open architecture and allows the programmer to distribute the intelligence between the single board computer (SBC) and PC compatible computer, which are connected through an interface. The SBC collects data from different sensors. System software resides on SBC and PC compatible computer. A PD controller has been designed and successfully implemented. Different new adaptive control algorithms, which require only link position measurements, were developed and tested, both on SUN workstation and on the experimental set-up.

Addresses two problems. One is to find the configuration-independent critical parameter for the articulated body inertia. The author analyzes the articulated body inertia matrix which appears in the discrete Riccati equation. It depends on kinematical parameters of the robot and on the inertial parameters of each link. Due to the dependence of the articulated body inertia on the kinematic parameters a standard discrete Fourier transform (DFT) is proposed to identify these parameters. Several examples show that this technique can be successfully applied to articulated inertia parameters. The remaining part of the paper is devoted to adaptive control algorithms for robots whose dynamics are described by making use of quasi-coordinates. Two set point control algorithms are presented and illustrated by simulation results.

Compares standard and diagonalized Lagrangian robot dynamics equations. This comparison is motivated by diagonalized equations of motion for multibody open kinematic chains with N degrees of freedom. These equations are derived by making use of new joint velocity variables, /spl nu/, which can be viewed as time-derivative of Lagrangian quasi-coordinates, similar to those of classical mechanics. A detailed computational complexity analysis of the standard and diagonalized Lagrangian robot dynamics algorithms is discussed in the paper. The results have been compared with those existing in the robotics literature. The author has conducted extensive numerical studies on diagonalized equations of motion and tried to establish some level of confidence with existing methods for dynamics calculations.