Showing papers by "Ivan Petrović published in 2003"

State estimation of an electronic throttle body

Abstract: Electronic throttle body (ETB) is a device used in cars to regulate air inflow into the motor's combustion system. Its good behavior is crucial for the superimposed engine speed control system. However, electronic throttle body is a highly nonlinear process, and its only measurable state is the throttle valve position measured by a cheap potentiometer of low resolution, resulting in significant quantization noise. In order to apply an advanced control strategy, all states should be usually available and the measurement noise should be reduced. With these two goals in mind we have implemented an extended Kalman filter (EKF), as a common solution for state estimation of nonlinear systems, and an unscented Kalman filter (UKF), which is a preferable solution when the process nonlinearities are very strong. Both filters are based on discrete time piece-wise affine process model which uses new friction model. By experimental tests on a real ETB it is shown that UKF gives better estimates of its state variables.

An approach to motion planning of indoor mobile robots

Abstract: In this paper we present a motion planning approach of indoor mobile robots based on integration of A* path planning algorithm and dynamic window local obstacle avoidance method. A simple and efficient procedure to the selection of appropriate motion commands based upon alignment of trajectories generated by the dynamic window module and the global geometric path is proposed. Global occupancy grid map is incrementally updated in on-line manner. The algorithm is verified in Saphira simulated environment for differential drive Pioneer 2DX mobile robot (manufacturer ActivMedia Robotics) using laser range sensor.

Application of extended Kalman filter for road condition estimation

Abstract: High quality estimation of tire-road friction forces has important role in many automotive control systems like anti-lock brake systems (ABS), traction control systems etc. For this purpose an extended Kalman filter augmented with integral term has been employed. A procedure for selecting appropriate integral gain has been proposed. The proposed estimator has been compared to the well-known passivity based state estimator.

Application of the RBF neural networks for tire-road friction force estimation

Abstract: This paper deals with the problem of the robust tire-road friction force estimation. Good information about friction force generated in contact between wheel and road has significant importance in many active safety systems in modern vehicles (anti-lock brake systems, traction control, vehicle dynamic systems, etc). Since state estimators are usually based on exact model of process, they are therefore limited by the model accuracy. A new estimation scheme based on RBF neural networks is proposed in this paper. The neural network is added to the estimator to compensate the effects of the friction model uncertainties to the estimation quality. An adaptation law for the neural network parameters is derived using Lyapunov stability analysis. The proposed state estimator provides accurate estimation of the tire-road friction force when fiction characteristic is only approximately known or even completely unknown. Quality of the estimation is examined through simulation using one wheel friction model. Simulation results suggest very fast compensation of the changes of the model parameters (< 150 ms) even when they vary in a wide range (changes of 100% and more). Possible drawback of proposed estimation scheme is the fact that neural network does not give the information what particular parameter has changed.

Improvements of Occupancy Grid Maps by Sonar Data Corrections

Abstract: In this paper we address one of the major problems of mobile robots navigation, the creation of a map from local sensor data collected as the robot moves around an unknown environment. Map building is the problem of generating models of robot environments from sensor data and can be often referred as a concurrent mapping and localization problem. That is to build a consistent map, the mobile robot has to know its pose. We present here three approaches to create occupancy grid maps from sonar's data and suggest a simple solution to improve the mapping quality in cases of irregular disposition of the sonars. The proposed solution has been tested on the mobile robot Pioneer 2DX.

Estimation of lube oil viscosities on a vacuum distillation column

Abstract: The key product specifications on a vacuum distillation column side draw products are viscosity, density, flash and color, although viscosity is usually the most limiting quality. There are online viscosity analyzers available on the market, but their prices are quite high and their reliabilities could be quit low. Possible solution to this problem is development and application of so-called soft sensors, which estimate lube oil viscosities based on available easy-to-measure variables. In this paper we consider the application of neural networks for viscosities estimation of lube oils on the vacuum unit at INA Rijeka lube oil refinery. The proposed soft sensors, based on neural networks, are of nonlinear finite impulse response structure, where their inputs are temperatures and flow-rates of the distillates. Although developed soft sensors do not demonstrate high accuracy they can be used to track trends of the output viscosities, and based on that information plant operators can take proper actions with more certainty. It is to expect some improvements in sensors behavior with arrival of new input/output data.