Sajith Akhil Vinayak, <br/><br/>
Path planning for autonomous navigation in roundabouts using an improved triangular based polynomial estimation technique for bezier curve generation with aa and aa* algorithms / 
Sajith Akhil Vinayak 
 - xiv, 117 pages :  illustrations (some color) ;  30 cm. +  1 CD ROM 
<br/><br/> Faculty of Manufacturing and Mechatronic Engineering Technology 
<br/><br/>Thesis (Master of Science) -- Universiti Malaysia Pahang – 2023 
<br/><br/>Includes bibliographical references 
<br/><br/> Autonomous vehicle (AVs) development is a rapidly growing research field. The AVs need to inspect the environment and detect the obstacle to create a path according to the environment. Path planning is a critical stage in the operation of vehicles, wherein the path is generated based on the surrounding environment to ensure safe and efficient navigation. The path is created for the vehicle to travel through different intersections. Among that roundabout is a type of intersection where path planning is challenging due to their complex shape and different traffic rules. To overcome this, different curve fitting methods were used to create path. However, the commonly used method is Bezier curve based curve fitting method which depends on the position of the control points. Many researchers have established different methods to position the control points but the change in the shape of roundabout cause the path to be inaccurate as the position of the control points are fixed. Therefore, the objective of the research is to introduce an enhanced method for calculating the control points for Bezier curve generation according to the shape of the roundabout. A triangular-based polynomial estimation technique is introduced, which helps calculate the Bezier curve's control points based on the points selected in the path. An equation is introduced which helps in calculating the control points based on the points selected and the respective segmentation factor (𝑡) value. The point selection in the roundabout is carried out by creating a triangular section where points are selected inside the triangle based on ratios (AA algorithm) or depending on the edge coordinates (AA star algorithm). The segmentation factor value for the points are calculated based on the distance between the points to avoid curve overfitting or underfitting. The circular path inside the roundabout is also created using the proposed method so that path can be adjusted according to the shape of the roundabout. The proposed method was tested in different roundabouts with different shapes to check the efficiency of the proposed method. The created path is tested using the Model predictive controller (MPC) vehicle model. Additionally, the proposed method is compared with other Bezier curve based path planning methods. The proposed method has RMS error of 0.208m for entry and 0.1319m for exit curve with the reference path which is the lowest compared to other path planning methods. The triangular-based polynomial estimation technique helps create the path according to the roundabout shape, thus overcoming the drawbacks of the previous methods. 
<br/><br/><label>ISBN: </label>THE0009645 (Local) 
<br/><br/><label>Subjects--Corporate Names: </label><br/>Faculty of Manufacturing and Mechatronic Engineering Technology--Dissertations
<br/><br/><label>Subjects--Topical Terms: </label><br/>Universities and colleges--Dissertations<br/>Theses--Dissertations