Publication: Optimización de la trayectoria del iCab basado en waypoints
Loading...
Identifiers
Publication date
2016-06
Defense date
2016-07-06
Authors
Tutors
Journal Title
Journal ISSN
Volume Title
Publisher
Impact
Export
Abstract
En el presente trabajo fin de grado se desarrolla un algoritmo de navegación para un vehículo autónomo. El algoritmo está basado en el método geométrico de “path tracking” Pure Pursuit. Los fundamentos teóricos del modelo Pure Pursuit, así como todos los problemas geométricos y otros detalles referentes a su implementación, se detallan en los capítulos 2 y 3 de este documento.
La técnica de navegación está basada en el uso de “waypoints”, es decir, el camino que debe seguir el vehículo está definido por una serie de puntos discretos en un plano XY. Normalmente los algoritmos de navegación no parten directamente de los waypoints, sino que consideran que ya existe una trayectoria dada.
El algoritmo desarrollado consta de dos partes diferenciadas. En primer lugar se crea la trayectoria a partir de los waypoints. Para lo cual es necesario aplicar algún método matemático de ajuste de curvas, es decir, encontrar una función “ f(x)” que se ajuste lo máximo posible a los “N” puntos seleccionados (waypoints).
En segundo lugar se aplica el modelo Pure Pursuit para calcular los comandos necesarios para la navegación autónoma del vehículo. Esto incluye el cálculo de un punto objetivo y el ángulo de dirección a aplicar para alcanzar dicho punto.
This bachelor dissertation shows the development of a navigation algorithm for an autonomous ground vehicle. The algorithm is based on the Pure Pursuit path tracking method. All the theoretical principles and the relevant geometric issues are explained in detail on chapters 2 and 3. The navigation technique used is based on waypoints. This means that the desired path, that the vehicle should follow, is defined by a distribution of points on a XY plane. Usually the navigation algorithms consider that there is already a given path to follow. However in this case the algorithm developed also creates a path using the waypoints. Therefor the algorithm is divided in two parts. First of all, a path is created using a curve fitting method. Mathematically that means finding a function “f(x) ” which best fits the “N” current waypoints used. Secondly, the Pure Pursuit method is used to calculate an achievable goal point on the defined path, and the necessary steering angle to command the vehicle.
This bachelor dissertation shows the development of a navigation algorithm for an autonomous ground vehicle. The algorithm is based on the Pure Pursuit path tracking method. All the theoretical principles and the relevant geometric issues are explained in detail on chapters 2 and 3. The navigation technique used is based on waypoints. This means that the desired path, that the vehicle should follow, is defined by a distribution of points on a XY plane. Usually the navigation algorithms consider that there is already a given path to follow. However in this case the algorithm developed also creates a path using the waypoints. Therefor the algorithm is divided in two parts. First of all, a path is created using a curve fitting method. Mathematically that means finding a function “f(x) ” which best fits the “N” current waypoints used. Secondly, the Pure Pursuit method is used to calculate an achievable goal point on the defined path, and the necessary steering angle to command the vehicle.
Description
Keywords
Tecnología automovilística, Algoritmos, Modelos matemáticos, Vehículos autónomos