This thesis work presents a model predictive control (MPC) based trajectory planner forhigh speed lane change and low speed parking scenarios of autonomous four wheel steering(4WS) vehicle. A four wheel steering vehicle has better low speed maneuverabilityand high speed stability compared with normal front wheel steering(FWS) vehicles. TheMPC optimal trajectory planner is formulated in a curvilinear coordinate frame (Frenetframe) minimizing the lateral deviation, heading error and velocity error in a kinematicdouble track model of a four wheel steering vehicle. Using the proposed trajectory planner,simulations show that a four wheel steering vehicle is able to track different type ofpath with lower lateral deviations, less heading error and shorter longitudinal distance. / I detta avhandlingsarbete presenteras en modellbaserad prediktiv kontroll (MPC) -baseradbanplaneringsplan f¨or h¨oghastighetsbanan och l°aghastighetsparametrar f¨or autonomtfyrhjulsdrift (4WS). Ett fyrhjulsdrivna fordon har b¨attre man¨ovrerbarhet med l°ag hastighetoch h¨oghastighetsstabilitet j¨amf¨ort med vanliga fr¨amre hjulstyrningar (FWS). MPC-optimalbanplanerare ¨ar formulerad i en kr¨okt koordinatram (Frenet-ram) som minimerar sidof¨orl¨angningen,kursfel och hastighetsfel i en kinematisk dubbelsp°armodell av ett fyrhjulsstyrda fordon.Med hj¨alp av den f¨oreslagna banaplaneraren visar simuleringar att ett fyrhjulsstyrfordonkan sp°ara olika typer av banor med l¨agre sidof¨orl¨angningar, mindre kursfel ochkortare l¨angsg°aende avst°and.

Tato diplomová práce se zabývá revizí elektroniky robotického vozidla car4 se zaměřením na měření rychlosti otáčení kol, výkonovou a řídicí elektroniku. Car4 posloužilo jako základ pro již přes 20 diplomových prací od roku 2010. To znamenalo, že některé základní aspekty car4 byly zastaralé. Bylo vytvořeno schéma elektroniky vozidla, které bylo dále využito pro vývoj a výrobu nového hardwaru. Komponenty byly poté otestovány a implementovány na vozidlo. Dále byl vytvořen kinematický model 4WS vozidla za použití Ackermanovy geometrie, který byl implenetován a otestován na car4. Dále by měl sloužit jako základ řídicího algoritmu pro budoucí vývoj car4.

This theasis deals with modeling dynamics and visualization of experimental vehicle CAR4 in software V-REP. The first part shows introduction with V-REP and its possibilities for usage in simulations. The second part presents modelling of CAR4 vehicle and its control by using software V-REP and Matlab. In last part are models of vehicle tested in simulation

[pt] Atualmente, as novas tecnologias estão estendendo os limites físicos dos veículos automotivos em busca de mais segurança e comforto. Novas aplicações, como por exemplo veículos autônomos, exigem sistemas de controle capazes de garantir a estabilidade do veículo durante a condução autônoma ou em cenários perigosos. Na maioria dos carros modernos, os sistemas de controle atuam de forma independente, ou seja, não há coordenação ou compartilhamento de dados entre eles, pois poderiam produzir conflitos entre esses controladores. Desse modo, nenhuma melhoria na estabilidade do veículo é alcançada ou inclusive, piores cenários podem ser produzidos. Para superar esses problemas, uma abordagem integrada é projetada neste trabalho. Esta integração, definida como sistema de controle integrado (IC), usa uma regra para coordenar o programa eletrônico de estabilidade (ESP em inglês) e o sistema de direção de quatro rodas (4WS em inglês). O ESP realiza uma frenagem seletiva dependendo do estado atual do veículo. Esta condição é estimada pela diferença entre a taxa de guinada desejada, obtida usando um modelo linear do veículo, e a taxa de guinada real. Adicionalmente, as pressões de frenagem em cada roda são calculadas pelo sistema de travagem antibloqueio (ABS em inglês). Neste trabalho, uma lógica de comutação on-off e um modelo hidráulico de primeira ordem são empregadas para modelar o sistema ABS. Para projetar o 4WS, usou-se uma estratégia por alimentação direta que considera o ângulo de esterçamento das roas frontais. Finalmente, para testar as vantagens do sistema IC proposto nesta tese contra o enfoque não integrado, realizaram-se simulações considerando um modelo não-linear do veículo em cenários críticos de condução. O modelo do veículo foi derivado empregando a abordagem multicorpos e o princípio de Jourdain, e depois é validado usando um conjunto de dados experimentais obtidos por sensores montados em um carro a escala. / [en] Nowadays new technologies are pushing the road vehicle limits further. Promising applications, e.g. self-driving cars, requires control systems that are able to ensure the vehicle s stability during autonomous driving or under dangerous scenarios. In most of modern cars, the control systems actuates independently, i.e. there is no coordination or data sharing between them. This approach can produce conflicts between these standalone controllers, thereby no improvements on the vehicle s stability are achieved or even a worse scenario can be produced. In order to overcome these problems, an integrated approach is designed in this work. This integration, defined as Integrated control system (IC), use a rule to coordinate the Electronic stability program (ESP) and the Four-wheel steering system (4WS). The ESP performs a selective braking depending of the current state of the vehicle. This condition is estimated by the difference between the desired yaw rate, obtained using a linear vehicle model, and the actual yaw rate. In addition, the braking pressures at each wheel are computed by the Anti-lock braking system (ABS). In this work, an on-off switching logic and a firstorder hydraulic model are employed to model the ABS system. To model the 4WS, a simple feed-forward control strategy that consider the front steering as input is used. Finally, in order to test the advantages of the IC system against the non-integrated one, simulations considering a nonlinear vehicle model under critical driving scenarios were performed. The vehicle model was derived employing the multibody approach and the Jourdain s principle, and then it is validated using a set of experimental data obtained by sensors mounted on a scaled car.

Four wheel steering (4WS) is a concept proven to be beneficial in low speed applications requiring large steering angles, which is the case in city traffic or parking. By steering the rear wheels in the opposite direction to the front ones, maneuverability can be improved. However, a conflict is encountered at high speeds for all the steering strategies developed. If sharper response is achieved, this is at the expense of undesirably large vehicle sideslip angles. On the other hand, small vehicle sideslip angles are associated with heavy understeering behavior. It is not possible to improve both simultaneously in case of two-axle 4WS vehicles. The object of this study is the simulation of various steering configurations for multi-axle vehicles in an attempt to find a means of solving the problem of 4WS and to determine the best steering strategy. In addition to two-axle vehicles which have been extensively studied in literature, three- and four-axle vehicles are taken into consideration. By extending the strategies used for 4WS two-axle vehicles, new strategies are established for three and four-axle vehicles. An integrated non-linear ride and handling model in Matlab &amp / Simulink environment considering sprung and unsprung mass motions, wheel and tire dynamics, is used for simulations. It is shown by case studies that, with the application of the derived strategies for three and fouraxle vehicles, lateral acceleration and yaw velocity responses can be improved without degrading vehicle sideslip angle.

The main aim of this research project is to extend theories of four-wheel-steering as developed by J. Ackermann to include an individually steered four-wheel steering system for passenger vehicles. Ackermann’s theories, including theories available in this subject area, dwell much on vehicle system dynamics developed from what is called single track model and some call it a bicycle model. In the bicycle model, the front two wheels are bundled together. Similarly, the rear wheels are bundled together. The problem with this is that it assumes two front wheels or two rear wheels to be under the same road, vehicle and operating conditions. The reality on the ground and experiments that are conducted are to the contrary. Therefore this study discusses vehicle disturbance rejection through robust decoupling of yaw and lateral motions of the passenger vehicle. A mathematical model was developed and simulated using Matlab R2008b. The model was developed in such a way that conditions can be easily changed and simulated. The model responded well to variations in road and vehicle conditions. Focus was in the ability of the vehicle to reject external disturbances. To generate yaw moment during braking, the brake on the left front wheel was disconnected. This was done because lateral wind generators, as used by Ackermann, were not available. The results from both simulations and experiments show disturbance rejection in the steady state. Copyright / Dissertation (MSc)--University of Pretoria, 2010. / Mechanical and Aeronautical Engineering / unrestricted