Papers by erhan ilhan Konukseven
This paper proposes radial basis function neural networks approach to the solution of a mobile ro... more This paper proposes radial basis function neural networks approach to the solution of a mobile robot heading adjustment using reinforcement learning. In order to control the heading of the mobile robot, the neural networks control system have been constructed and implemented. Neural controller has been charged to enhance the control system by adding some degrees of strength. It has been achieved that neural networks system can learn the relationship between the desired directional heading and the error position of the mobile robot. The radial basis function neural networks have been trained via reinforcement learning function approach. The performance of the proposed controller and learning system has been investigated by using mobile robot that consists of a two driving wheels mounted on the same axis, and a front passive wheel for balance.
The International Journal of Advanced Manufacturing Technology, Dec 24, 2016
All information in this document has been obtained and presented in accordance with academic rule... more All information in this document has been obtained and presented in accordance with academic rules and ethical conduct. I also declare that, as required by these rules and conduct, I have fully cited and referenced all material and results that are not original to this work.
Springer eBooks, 2010
ABSTRACT The aim of this research was to design an optimum 6R passive haptic robotic arm (PHRA) t... more ABSTRACT The aim of this research was to design an optimum 6R passive haptic robotic arm (PHRA) to work in a limited workspace during dental implant surgery. Misplacement of an implant during dental surgery causes longer recuperation periods and functional disorders. In this study, a passive guidance robot arm was designed as a surgical aid tool for a dentist during the operation to reduce the surgical complications. Optimum design of a 6R robot is a complex issue since minimum energy has to be consumed while maximum workspace is to be achieved using optimized link lengths. The methodology used deals not only with link lengths of manipulator but also mass and inertia of the links along with the location of the tool path. Another feature of the methodology is to maximize haptic device transparency using an objective function that includes end-effector torques/forces with workspace limits taken as constraints. The objective function was obtained from dynamic equations and the constraints were defined using kinematic equations. The constrained nonlinear optimization problem was solved using Sequential Quadratic Programming (SQP) and Genetic Algorithms (GA). Main contribution of this paper is an optimization algorithm that considers spatial dynamics to reduce parasitic torques leading to an optimal 6R robot design. Details of the methodology, solutions, and performance of the optimization techniques are presented.
Biosystems Engineering, Jun 1, 2016
In this paper, the effects of wheel slip estimation and compensation of trajectory tracking for o... more In this paper, the effects of wheel slip estimation and compensation of trajectory tracking for orchard applications were investigated. A slippage estimator was developed and adapted into a car-like robot model. Steering and velocity commands were generated using a model-based control approach. The whole system was implemented and tested on an autonomous orchard vehicle that has steerable front wheels and actuated rear wheels. A high accuracy positioning system was used to estimate the longitudinal and lateral slip velocities while the vehicle is moving. A laser scanning range finder was placed at the front centre of the vehicle, which was used to detect rows of trees in the orchard. Procedures were first tested in a non-flat but open space, which was covered with snow. Then it was tested on an experimental orchard where the surface was covered with heavy mud and the vehicle was expected to follow trajectories that span multiple rows in the orchard. The vehicle detected individual trees as well as rows of trees to track the centre of each row and manoeuvred from one row to the next. The experimental results showed that trajectory tracking performance of the vehicle was enhanced via integrating a slippage estimator into the system model. Furthermore, using the slippage estimation in the system model increased the accuracy, repeatability and performance of the control system. Keywords Autonomous orchard vehicle; Slippage estimation; Row following; Turning. Nomenclature RTK-GPS real time kinematic-Global Positioning System WD wheel drive ROS Robot Operating System x, y coordinate system in the centre of the rear axle of the vehicle X, Y vehicle's current position
Mechanism and Machine Theory, Jun 1, 2010
The reduction of the pressure in the compression volume of a McLeod manometer, due to the mercury... more The reduction of the pressure in the compression volume of a McLeod manometer, due to the mercury stream from the reservoir to the cold trap, has been further in vestigated. The pressure reading for different gases by a McLeod manometer was compared with that by an ionization manometer. The gases studied were neon, argon, krypton, xenon and nitrogen. Pressures were varied by a factor of 5 between 10-4 and 10-3 torr. The temperature was varied between 0 and 40 °C. Two different diameters of the gas tube which connects the McLeod compression volume with the cold trap, were tested. It was concluded that a simple calculation of the real pressure from the McLeod manometer reading is possible. The hard sphere approximation often used for the calculation of the diffusion coefficients is not sufficiently accurate, due to the considerable attractive forces between Hg atoms and the gas atoms used.
Presence: Teleoperators & Virtual Environments, Jun 1, 2011
Position, velocity and acceleration information are important for mobile robots. Due to the wheel... more Position, velocity and acceleration information are important for mobile robots. Due to the wheel slippages, encoder data may not be reliable and IMU data also contains a cumulative error. Errors of inertial measurements are accumulated over velocity and position estimates and as time increases, these errors grow higher. Due to robot hardware and the operating surface, ground truth may not be available. In this work recurrent deep neural network is proposed in order to reduce the error in speed and yaw angle estimates coming from encoder and IMU data. Neural networks are commonly used to capture the behavior of linear and nonlinear systems. Since ground-wheel interaction forces are modeled with non-linear models such as the Magic formula and determining parameters of those models require time and test setups, there is a need for simpler methods to model the behavior of simple robots. Neural networks could be used to model non-linear systems. In this work, a recurrent deep neural network is proposed to estimate the speed and yaw angle of a two-wheeled differentially driven mobile robot. Using the information coming from the camera positioned above the test area as ground truth, the network is trained. After that, the output of the network is recorded in the absence of ground truth information in the network. Finally, the performance of the network is evaluated using network output, sensor data calculation, and ground truth.
Lecture Notes in Computer Science, 2012
Transparency of a haptic interface can be improved by minimizing the effects of inertia and frict... more Transparency of a haptic interface can be improved by minimizing the effects of inertia and friction through the use of model based compensators. However, the performance with these algorithms is limited due to the estimation errors in the system model and in the velocity and acceleration from quantized encoder data. This paper contributes a new torque compensator based on motor current to improve transparency. The proposed method was tested experimentally in time and frequency domains by means of an excitation motor attached at the user side of the device. The excitation motor enabled evaluation of the algorithms with smooth trajectories and high frequencies, which cannot be generated by user hand. Experimental results showed that the algorithm significantly improves transparency and doubles the transparency bandwidth.
Mechanism and Machine Theory, Jun 1, 2012
ABSTRACT
AIAA SCITECH 2023 Forum, Jan 19, 2023
Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science, Sep 13, 2012
The literature on kinematic calibration of industrial robots and haptic devices suggests that pro... more The literature on kinematic calibration of industrial robots and haptic devices suggests that proper model calibration is indispensable for accurate pose estimation and precise force control. Despite the variety of studies in the literature, the effects of transmission errors on positioning accuracy or the enhancement of force control by kinematic calibration is not fully studied. In this article, an easy to implement kinematic calibration method is proposed for the systems having transmission errors. The presented method is assessed on a 7-DOF Phantom-like haptic device where transmission errors are inherently present due to the use of capstan drives. Simulation results on pose estimation accuracy and force control precision are backed up by experiments.
In this paper, We have extracted and modeled the motion primitive of human hand in a simple 1-DOF... more In this paper, We have extracted and modeled the motion primitive of human hand in a simple 1-DOF rhythmic motion task, i.e. manipulating mass-spring-damper system. The experiment was carried out by using 1-DOF haptic box with virtual reality in Simulink environment. The interaction dynamics of haptic box and human which consists of hand and brain reveals the role of the human as an intelligent admittance. We tested 6 people who tried to combine motion primitives to produce smoother motion during learning process. In addition, we developed a novel identification method for modeling the rhythmic motion of hand in model space. It is shown that adaptive filter as a predictor of motion primitives with two parameters and two initial values appears as an ellipse in model space. The geometrical properties of ellipse are related to the parameters and initial values of adaptive filter that make it possible to identify the parameters of adaptive filter in model space.
International Journal of Design Engineering, 2014
In an efficient autonomous navigation and exploration, the robots should sense the environment as... more In an efficient autonomous navigation and exploration, the robots should sense the environment as exactly as possible in real-time and act correctly on the basis of the acquired 3D data. In this paper, we present the design and implementation of a scanning platform, which can be used for both outdoor and indoor mobile robot navigation and mapping. A 3D scanning platform based on a 2D laser rangefinder was designed in a compact way with a maximum field of view for fast and accurate mapping. The designed mechanism provides 360° degree horizontal by 240° degree vertical field of view. The maximum resolution is 0.36° degrees in elevation and variable in azimuth. The proposed low cost compact design is tested by scanning a physical environment with known dimensions to show that it can be used as a precise and reliable high quality 3D sensor for autonomous mobile robots.
Robotica, Jun 3, 2014
During the past decades, intensive research has been pursued on the development of kinetic models... more During the past decades, intensive research has been pursued on the development of kinetic models to predict process behavior in ethanol production from lignocellulose. These models comprise a large number of parameters which have to be tuned with appropriate experimental data. Therefore, the parameter estimation problem plays an essential role. This work addresses the parameter estimation problem in models representing dilute acid hydrolysis, detoxification, and cofermentation operations in the biochemical production of ethanol from lignocellulosic biomass. The models are represented by sets of differential-algebraic equations (DAEs). Unlike previous approaches, these models account for the main process variables that affect the entire process, specially the final production of bioethanol. These detailed kinetic models, systematically tuned with experimental data, can be used in future studies within a model-based framework that allows performing realistic simulation and optimization aimed at bioethanol process design. A sensitivity analysis has been performed in order to identify the most sensitive parameters. The parameter estimation problem is solved with a simultaneous optimization approach in which the system of dynamic equations is converted into a set of algebraic ones through orthogonal collocation on finite elements. Thus, estimating the model parameters entails optimizing a weighted least squares objective function subject to the discretized algebraic constraints, resulting in a large-scale nonlinear programming problem (NLP). A good agreement with available experimental data has been obtained with estimated kinetic parameters in each model.
Makina Tasarım ve İmalat Dergisi, May 1, 2017
Uploads
Papers by erhan ilhan Konukseven