This paper proposes a novel fuzzy-adaptive extended Kalman filter (FAEKF) for the real-time attitude estimation of agile mobile platforms equipped with magnetic, angular rate, and gravity (MARG) sensor arrays. The filter structure employs both a quaternion-based EKF and an adaptive extension, in which novel measurement methods are used to calculate the magnitudes of system vibrations, external accelerations, and magnetic distortions. These magnitudes, as external disturbances, are incorporated into a sophisticated fuzzy inference machine, which executes fuzzy IF-THEN rules-based adaption laws to consistently modify the noise covariance matrices of the filter, thereby providing accurate and robust attitude results. A six-degrees of freedom (6 DOF) test bench is designed for filter performance evaluation, which executes various dynamic behaviors and enables measurement of the true attitude angles (ground truth) along with the raw MARG sensor data. The tuning of filter parameters is performed with numerical optimization based on the collected measurements from the test environment. A comprehensive analysis highlights that the proposed adaptive strategy significantly improves the attitude estimation quality. Moreover, the filter structure successfully rejects the effects of both slow and fast external perturbations. The FAEKF can be applied to any mobile system in which attitude estimation is necessary for localization and external disturbances greatly influence the filter accuracy.

The Szabad(ka)-II hexapod robot with 18 DOF is a suitable mechatronic device for the development of hexapod walking algorithm and engine control [1, 2]. The required full dynamic model has already been built [3], which is used as a black-box for the walking optimizations in this research. The ellipse-based walking trajectory has been generated that was required by the low-cost straight line walking [4], and the purpose was to optimize its parameters. The Particle Swarm Optimization (PSO) method was chosen for simple and effective working, which does not require the model's mathematical description or differentiation. Previously the authors performed an evolutionary Genetic Algorithm (GA) optimization for a similar trial case [5], and posed the principles of the quality measurement of hexapod walking [4, 5]. The same visual evaluation and comparison was applied in this paper for the results of both optimization methods. PSO has produced better and faster results compared to GA.

A simulation model contributes to the development of both algorithms based on navigation sensors and their application in real nonlinear mechatronic systems. The Szabad(ka)-II hexapod walker robot is equipped with an inertial measurement unit (IMU), and this paper presents a novel calibration procedure of its simulation model. Various sinusoidal calibration movements were performed on both the model and the robot, and the raw IMU measurements were recorded simultaneously with motor electrical parameters and joint movement variables. The simulation model includes the model of IMU sensors, where the location, misalignments, and scaling parameters are also incorporated in the tuning procedure. Thus, this simulation environment enabled the calibration procedure to be performed based on the measurement data. The efficient optimization of both the unknown and estimated parameters of the robot model along with the IMU sensor model resulted in a simulation output that fits the measurement results satisfactorily. The nominal and remaining errors were analyzed both statistically and in the spectral domain. Due to the proposed method, the simulation error of the accelerometer and gyroscope measurements were decreased by 35%. The necessity of calibrating the sensor model was justified via the application of an extended Kalman filter (EKF) for the attitude estimation.

The Szabad(ka)-II 18 DOF walking robot and its simulation model is suitable for research into hexapod walking algorithm and motion control. The complete dynamic model has already been built, and is used as a black box for walking optimization in this research. First, optimal straight line walking was chosen as our objective, since the robot mainly moves in this mode. This case can be tested and validated as well on the current version of our robot. An ellipse-based leg trajectory has been generated for this low-cost straight line walking. Currently a simple new Fuzzy-PI controller with three input variables is being constructed and compared with an previously used PI controller. The purpose of defining the rules and its optimization are to obtain a controller that provides walking with higher quality. Both the compared controllers have been optimized together with the parameters of the leg trajectory. The particle swarm optimization (PSO) method was chosen from several methods with our benchmark-based selection research and the help of specific test functions; moreover the previous research (the comparison of genetic algorithm (GA) and PSO) also led to this conclusion.

The utilization of stereo cameras in robotic applications is presented in this paper. The use of a stereo depth sensor is a principal step in robotics applications, since it is the first step in sequences of robotic actions where the intent is to detect and extract windows and obstacles that are not meant to be painted from the surrounding wall. A RealSense D435 stereo camera was used for surface recording via a real-time, appearance-based (RTAB) mapping procedure, as well as to navigate the painting robot. Later, wall detection and the obstacle avoidance processes were performed using statistical filtering and a random sample consensus model (RANSAC) algorithm.

SUMMARY: Our complete dynamical simulation-model realistically describes the real low-cost hexapod walker robot Szabad(ka)-II within prescribed tolerances under nominal load conditions. This validated model is novel, described in detail, for it includes in a single study: (a) digital controllers, (b) gearheads and DC motors, (c) 3D kinematics and dynamics of 18 Degree of Freedom (DOF) structure, (d) ground contact for even ground, (e) sensors and battery model. In our model validation: (a) kinematical-, dynamical- and digital controller variables were simultaneously compared, (b) differences of measured and simulated curves were quantified and qualified, (c) unknown model parameters were estimated by comparing real measurements with simulation results and applying adequate optimization procedures. The model validation helps identifying both model's and real robot's imperfections: (a) gearlash of the joints, (b) imperfection of approximate ground contact model, (c) lack of gearhead's internal non-linear friction in the model. Modeling and model validation resulted in more stable robot which performed better than its predecessors in terms of locomotion.