Mateusz Przybyla

Application of Active Disturbance Rejection Control to a reel-to-reel system seen in tire industry

In this paper an Active Disturbance Rejection Control (ADRC) scheme is used to solve a sag control problem that has been noticed in tire production process. Constant length of the rubber sag is crucial in order to achieve high quality of tire reinforcement, which has its direct impact on driver safety. The ADRC method was compared in this research with other model-independent controller, i.e. Proportional-Integral-Derivative (PID), widely used in engineering practice. Both of the techniques were implemented and experimentally verified on a model of tire reinforcement production stage that can be described with a reel-to-reel system (R2R). The effect of various radius in the rubber winding phase makes the R2R a nonstationary dynamic plant. The ADRC approach resulted in better performance both energetically and for tracking a square trajectory, even with the presence of artificially added external disturbance. Additional perturbation imitated unpredictable and surprisingly often phenomena, which are results of incidents that happen in other stages of the manufacturing process. The acquired results have potential importance to the tire industry.

Active Disturbance Rejection Control for a Flexible-Joint Manipulator

This paper focuses on experimental verification of Active Disturbance Rejection Control (ADRC) on a one-link flexible-joint manipulator. The complexity of the considered system limits the ease of modeling and thus the performance of model-based control methods. ADRC is proposed in this research as an alternative to these techniques. It uses a disturbance observer to actively compensate the effects of perturbations acting on the system. A set of experiments was conducted in order to examine the robustness of the proposed control framework. The results obtained show that the ADRC method managed to stay robust against nonlinear behavior of the flexible-joint system as well as its dynamics parameters variations.

Set-point Control of Mobile Robot with Obstacle Detection and Avoidance Using Navigation Function - Experimental Verification

This paper presents the results of an experimental verification of mobile robot control algorithm including obstacle detection and avoidance. The controller is based on the navigation potential function that was proposed in work (Urakubo, Nonlinear Dyn. 81(3), 1475–1487 2015). Conducted experiments considered the task of reaching and stabilization of robot in point. The navigation potential agregates information of robot position and orientation but also the repelling potentials of obstacles. The obstacle detection is performed solely with the use of laser scanner. The experiments show that the method can easily handle environments with one or two obstacles even if they instantly hide or show-up due to the scanner range limits. The experiments also indicate that the utilized control method has a good potential for being used in parallel parking task.

Control of a mobile robot and collision avoidance using navigation function - experimental verification

In this paper experimental verification of a mobile robot control and colision avoidance is presented. The control is based on the navigation function that was proposed in [2]. Navigation function agregates position, orientation and colision avoidance terms. Experimental results show that both position and orientation lead to desired values very quickly and reach them with small errors.

Saddle point detection of the navigation function in nonholonomic mobile robot control

This paper presents navigation function saddle point detection and avoidance for a unicycle robot in sphere worlds. The detection is based on the gradient and Hessian analysis. The avoidance algorithm was numerically tested for two scenarios and various values of the parameter connected with the sensitivity of the saddle point detection.

An experimental comparison of model-free control methods in a nonlinear manipulator

This paper presents an experimental comparison of various model-independent control strategies implemented on a system with unbalanced rotating mass. Proportional-Integral-Derivative (PID), Robust Tracking with Control Vector Constraints (RTCVC), and Active Disturbance Rejection Control (ADRC) are considered in this research. Although these control algorithms deal with parametric uncertainties, external disturbances, and nonlinear behavior of the system with different approaches, their common feature is that they do not need an explicit mathematical model of the physical process. Obtained results show that the ADRC achieved highest control performance in terms of position trajectory tracking of the manipulator link and energy efficiency. This work also confirms that the ADRC is a promising model-free control approach, which brings together what is best in both classic and modern control theories.

Rapid navigation function control for omnidirectional mobile platform

This paper presents an extension of navigation function used to control an omnidirectional robot. Navigation function is used to control position coordinates while the orientation variable is controlled with simple proportional controller. The extension relies on a specific normalization of navigation function gradient. Presented method results in much more rapid convergence in comparison to classic approach based on negative gradient of the navigation function. The most noticeable result of the extension is observed for high values of κ parameter, which must be increased if the distances between obstacles are small. Experimental results are given to illustrate effectiveness of the proposed algorithm.

Detection and tracking of 2D geometric obstacles from LRF data

This work proposes a method for detection and tracking of local geometrical obstacles from sequences of two-dimensional range scans. Detected obstacles are represented with linear or circular models. Circular obstacles are subject to tracking algorithm based on Kalman filter. Solutions to both the correspondence and the update problem of the tracking system are provided. The occurence of obstacles fusion or fission is defined and addressed. A by-product of the system is the information on tracked obstacles velocity. The algorithm is dedicated to wheeled mobile robots with mounted planar laser range finders.

Reactive obstacle avoidance in crowded environments for 2D omni-directional robot

The work describes a heuristic control strategy comprising reactive obstacle avoidance for omni-directional planar robot. The proposed method is based on artificial potential fields. Appropriate reshaping of repelling potential fields incorporating warping vectors and warping functions reduces the problem of local minima. Moreover, a solution to saddle point avoidance based on potential hessian is proposed. Numerical tests are conducted to indicate the validity and efficiency of the method. The strategy applies only to circular non-overlapping obstacles.

Strategy for designing a control system for a target-approach task by a mobile robot

This paper presents a strategy for designing a control system for the mobile robots which task is to approach a target from a specific direction. The strategy is based on the concept of virtual forces combined with actual, physical forces acting upon the system. The virtual potential force field is a concatenation of a spring force and a dipol-like force. An additional braking force, based on work-energy theorem, is introduced in order to obtain virtual damping effect. Because of force-based approach, the strategy focuses on dynamical systems. The concept has been tested via simulations.