Examensarbeten för masterexamen // Master Theses
Länka till denna samling:
Browse
Browsar Examensarbeten för masterexamen // Master Theses efter Ämnesord "Control Engineering"
Visar 1 - 3 av 3
Sökresultat per sida
Sortera efter
- PostMap-Aided Vehicle Tracking Used in Behaviroural Studies for Intersection Driver Assistance Applications(2007) Haghighi, Kasra; Bagheri, Toktam; Chalmers tekniska högskola / Institutionen för signaler och system; Chalmers University of Technology / Department of Signals and SystemsThis thesis addresses the vehicle tracking problem in video recordings in intersections for driver behavior studies with the assistance of digital maps. For object detection and tracking, both the own-vehicle and other vehicle motion parameters like position, speed, heading and acceleration are needed. This information is measured using sensors installed on a test vehicle. These include GPS, gyros, video cameras, and a digital map. The digital map is used as a sensor for improving vehicle tracking performance. Since, all sensors contain errors and uncertainties, error models are computed for some certain sensors. To compare different tracking strategies, performance measures are introduced. By using the manual object selection (from a Graphical User Interface) in a few video frames, at least in the beginning and end, an estimation algorithm computes this vehicle's position along the road. This map estimation technique was enhanced by adding an image processing system, which looks for the previously selected object in the next frame. In order to include the other vehicle's motion and sensors error model a particle filter based tracking algorithm is designed. This method increases the tracking performance in the case that only a few manual markings have been given. It is shown in this study that with only a few manual markings, then the tracking/fusion algorithm will increase the performance by exploiting visual features from images. On the other hand, when there are many manual markings available, this simple image analysis does not extract more information than the map estimator.
- PostModello per la stima di coppia nei motori diesel(2003) Falcone, Paolo; Chalmers tekniska högskola / Institutionen för signaler och system; Chalmers University of Technology / Department of Signals and Systems
- PostVehicle modelling and washout filter tuning for the Chalmers Vehicle Simulator(2007) Murgovski, Nikolce; Chalmers tekniska högskola / Institutionen för signaler och system; Chalmers University of Technology / Department of Signals and SystemsThe Chalmers Vehicle Simulator (CVS) was built by students in 1999 and is constantly being upgraded ever since. The main objective is to provide a realistic simulation environment for students to perform projects and master thesis works, investigating and testing new products and ideas relevant to the car industry. It consists of a hexapod motion platform (Stewart platform) and five computers responsible for the simulation. A quarter of a Volvo car (the part where the driver sits) is mounted on the platform and the visual cues are projected on a screen in front of the driver. The algorithm that transforms the desired vehicle motion into realizable simulator motion commands is called a washout filter. The washout filter is responsible for keeping the motion platform within its physical boundaries and for stimulating the driver to feel that driving the simulator is close to driving a real car. The washout filter “washes out” cues below the driver’s perception threshold and returns the platform state to the neutral position. It calculates the platform position and angular displacement in real time, taking the desired translational acceleration and angular displacement as an input signal. Washout filters have been widely investigated, mainly in the field of flight simulations. In this work three washout filters originally developed for NASA airplane simulators are considered for the CVS. The Classical and Optimal washout filters are implemented for real-time use, while the Adaptive washout filter is tested only by off-line simulation. The quality of a washout filter depends on how realistic motion it produces. This goal is achieved by minimizing the difference between the measured sensed accelerations and rotations that a driver feels in a real car, with those the driver feels in the simulator. The sensed accelerations and rotations are determined by a mathematical model of the human vestibular system which is mainly responsible for motion sensations. An interesting feature of the vestibular system is that it does not differentiate between accelerations produced by translational movement and accelerations produced by tilting the driver’s head with respect to the gravity vector. This phenomenon is known as “tilt coordination” and makes it possible to simulate low frequency translational accelerations by tilting the platform. This augments the high frequency acceleration cues produced by the washout filters. The washout filter parameters are tuned by optimization algorithms. A Genetic Algorithm is used to find a starting point in the parameter space for the ensuing local optimization where a Riccati Algebraic Solver and the Steepest Descent Method are used. The optimization is performed on a computer simulation model of the CVS, taking standard driving maneuvers as inputs. The obtained Classical and Optimal washout filters were tested in real time on the CVS with several “test drivers”. During all the tests, the platform never hit the physical boundaries, but moved very close to them, thus using most of the actuator’s movement. According to the test drivers’ suggestions some of the input signals were rescaled. After the final adjustments their impression was that the washout filters produced realistic driving experience.