Goals and Objectives

Other results

• The following software programs for PCs and robotic systems have been developed: a program for plotting frequency response of sensitivity functions, a program for controlling mobile robots using gestures, voice, and facial expressions, a program for graphical analysis of transient processes in first- and second-order linear dynamic systems, a stabilization program for the Darwin-OP robot, a motion control program for the Darwin-OP robot using computer vision, a control program for the Pololu Dual VNH5019 driver, the Surfer program for estimating the position and stabilization of a humanoid robot on an inclined surface, a program for estimating the angular position of a humanoid robot and calculating the center of mass coordinates of its links, and a program for decentralized control of a rotorcraft.
• A test bench for vehicle suspension testing has been developed.
• Robust stabilization systems for the Snowboarder and Snowboarder 2.1 bipedal robots in a standing position have been developed.
• Modeling of a serial compensator under quantization conditions has been developed.
• A mobile robot for horizontal and vertical movement, a mobile robot, and a robotic gripper have been developed.
• A motion trajectory planning algorithm for a humanoid robot balancing in a standing position has been developed, software-implemented, and experimentally tested.
• A motion trajectory construction algorithm for a humanoid robot based on computer vision system data has been developed, software-implemented, and experimentally tested.

Scientific Results

• A simple output-based adaptive control algorithm has been developed for a class of linear perturbed systems with uncertainties. Its main advantage lies in its structural simplicity. The control law includes one adjustable parameter and has a low dynamic order. It guarantees closed-system robustness and exponential convergence of the control error to near-zero.
• An adaptive control algorithm for multi-input multi-output (MIMO) objects under parametric uncertainty has been developed. The algorithm was analyzed for a two-channel system, and conditions for the applicability of the control law were proposed to ensure exponential convergence of the object output norm to zero. Computer modeling was conducted for an unstable two-channel system, confirming the effectiveness of the proposed algorithm.
• A consensus control algorithm for a dynamic network consisting of four nodes has been constructed, where each node is described by a nonlinear differential equation with delay and unknown parameters. The algorithm ensures network synchronization and compensation of uncontrolled disturbances with a specified accuracy.
• An adaptive and robust control algorithm based on the sequential compensator method has been synthesized for controlling systems with parametric and structural uncertainties (in cases where only the maximum possible relative degree of the control object is known). To illustrate the algorithm's performance, a mobile robot motion control program using a computer vision system was developed.
• A stabilization algorithm has been developed that compensates for delay in a class of nonlinear systems and ensures asymptotic stability for a closed-loop nonlinear system under conditions of a biased sinusoidal disturbance signal. The constructed adaptive scheme allows for determining the frequency and other parameters of the external disturbance, which are used in the compensation loop.
• A new approach has been discovered for stabilizing unstable linear systems with large input delays, unknown parameters, and disturbances. A predictor-based algorithm is proposed that ensures system parameter identification and stabilization. An extended problem involving the estimation and compensation of an unknown disturbance was also analyzed. Numerical modeling was conducted to demonstrate the performance and efficiency of the proposed adaptive control.
• An output control method is proposed for a class of nonlinear MIMO (multiple-input, multiple-output) systems. For example, quadcopter control was implemented based on the sequential compensator method using mathematical model decomposition into two parts: the first system is a static MIMO transformation (the object is represented by a system of linear equations relating to motor lift forces and virtual control inputs), and the second is represented as several SISO (single-input, single-output) channels.
• A method for minimizing data flows when tracking a moving object via a digital communication channel based on a binary adaptive encoder is presented. The tracking of a moving unmanned aerial vehicle trajectory was considered as an example. A correlation was established between the communication channel load (amount of data transmitted per unit of time) and the accuracy of the tracking process.
• A system for estimating the speed of a vehicle moving along a road has been developed, based on measurements from a roadside sensor node including an accelerometer and a magnetometer. Proving ground test results showed high effectiveness of the proposed method, though further refinement is required for use on public roads.
• The problem of improving frequency identification for a single-frequency harmonic signal has been solved by constructing a filter cascade. The proposed cascade consists of adaptive bandpass filters tuned to frequency estimation provided by the proposed identification algorithm. The stability of the cascade was studied, and trajectory boundedness was proven through Lyapunov function analysis with the derivation of assumptions for the identification algorithm.
• A phase regulator for a population of oscillatory systems has been developed. The proposed approach is based on the Phase Response Curve (PRC) model for an isolated oscillator (a reduced first-order model obtained for a system linearized along the limit cycle with an infinitesimal input). It is proven that phase regulation is also achieved for the original nonlinear system.
• A new robust, nonlinear, globally convergent position observer for a fixed permanent magnet synchronous motor is proposed. A key feature of the proposed regulator is that it only requires knowledge of stator resistance and inductance, while mechanical parameters and magnetic flux can remain unknown. Due to a new reparameterization of the motor dynamics, only two parameters are estimated by the regressor, including filtered voltage and current, during normal motor operation.
• A new class of estimation devices for permanent magnet synchronous motors is proposed. By using a new representation of motor dynamics and suitable filters, a new solution was obtained for two important problems: estimating stator resistance and inductance, and estimating flux with unknown electrical parameters. Neither scheme requires knowledge of mechanical parameters or magnetic flux.
• A finite-time control method for a chain of integrators has been developed. The proposed approach ensures finite-time stability and robustness against external disturbances.
• ISS-stability conditions for systems with multiple equilibrium positions and time delays were investigated based on the Lyapunov-Razumikhin functional. The robustness of stability with respect to delays in the feedback channel was demonstrated.
• A control algorithm for significantly nonlinear second-order systems under conditions of parametric uncertainty and external bounded disturbances has been developed. The algorithm is built on backstepping methods and an auxiliary loop. This ensured system stability compared to the application of the classical backstepping method.
• A modified finite-time control algorithm has been obtained, based on the application of the implicit Lyapunov function method and the properties of homogeneous systems.
• An algorithm for solving the signal uncertainty problem in the analytical design of a serial compensator in a piezo drive control task was analyzed to verify its performance and efficiency.
• A method for synthesizing sensorless control algorithms for a non-salient pole permanent magnet synchronous motor using developed position observers was developed. The control system combined with a nonlinear observer was compared with a modern industrial controller. It was proven that the proposed sensorless controller with a nonlinear observer significantly improves closed-loop performance during motor operation at both low and high speeds.
• An adaptive tracking system for a multi-sinusoidal signal under conditions of input delay caused by using the Internet as a communication channel was tested to verify its operability and efficiency.
• Robust control algorithms for non-linear systems were developed, which, in the presence of delay in the control channel, ensure the convergence of all closed-loop system trajectories into a compact set. In the absence of delay, finite stability of the closed-loop system is guaranteed. The results obtained are also valid for cases of variable delay and multiple delays.
• The problem of output regulation for multi-channel systems with harmonic external inputs and parametric uncertainties was considered. The problem of robust stability using the "robust" minimum phase property was examined. A controller based on the classical internal model method combined with an adaptive tuning loop was developed, ensuring asymptotic convergence of output variables to equilibrium positions. Experimental testing of the developed algorithm for surface vessel control was conducted.
• An algorithm has been developed that allows extracting useful information from a chaotic signal to estimate system parameters. An observer was also created that uses only the output signal of the chaotic system under conditions of complete parametric model uncertainty.
• A path planning algorithm for industrial robots has been developed. It is based on approximating a given path using arcs. The solution reduces the number of waypoints, code size, and computational costs, improves operation quality, and simplifies the programming of complex movements. Experimental studies were conducted using a six-axis manipulator with revolute joints.
• An algorithm for estimating unknown solar cell parameters using the iterative Newton-Raphson method was developed.
• A method for identifying the maximum power point of a photovoltaic source under various atmospheric conditions based on P&O and INC methods has been proposed.
• A method for identifying solar panel parameters using a dynamic regressor extension procedure has been developed.

Education and personnel retraining

• Four Master's programs were created and implemented: "Digital Control in Modern Engineering" (2014), "Industrial Robotics" (2016), "Adaptive and Robust Control of Nonlinear Systems" (2017), "Sensorless Control" (2018), as well as a postgraduate program for research and teaching staff "Modern Technologies for Control Systems Synthesis" (2015).
• Defenses: 3 doctoral dissertations, 12 candidate dissertations, 25 Master's theses, 8 Bachelor's theses.
• The textbook "Designing Intelligent Control Systems for Home Automation. Elements of Theory and Workshop" has been published.
• In the Laboratory, 10 young scientists, specialists, and teachers from the University of Leeds (UK), the V. A. Trapeznikov Institute of Control Sciences of the RAS (Russia), Astrakhan State Technical University (Russia), Gubkin Russian State University of Oil and Gas (Russia), and Peter the Great St. Petersburg Polytechnic University (Russia) underwent professional retraining/advanced training in the "Adaptive and Nonlinear Control Systems" course.

Cooperation