A Bezier curve optimization method based on segmentation factor (t) for path planning in autonomous navigation

• Main Authors: Vinayak, Akhil, Zakaria, Muhammad Aizzat, Baarath, K.
• Format: Conference Proceeding
• Language: English
• Subjects: Autonomous navigation; Curvature; Curve fitting; Curves; Mathematical analysis; Path planning; Segmentation
• ISSN: 0094-243X; 1551-7616
• DOI: 10.1063/5.0188927

Path planning plays a major role in autonomous navigation. Different curve fitting methods are used in creating path for autonomous vehicles. Among those Bezier curves are widely used to create path especially in roundabouts due to their special properties. The primary factor that affects the shape and curvature of the Bezier curve is the control points. Researchers have established equations for calculating the control points for a Bezier curve travelling through selected points. The two critical elements that determine the control points are the random points selected for control point calculation and the segmentation factor (t) for each of the points selected. The values of segmentation factor (t) is assigned randomly to each point. As the segmentation factor change new control points are calculated which makes a change in the shape of the curve but the curve will pass through the selected points. So, this paper presents the change in the shape of the curve as the segmentation factor (t) changes. Additionally, it demonstrates the change in segmentation factor (t) for each point selected to fit the curve within the required space and to adjust t value for connecting two different Bezier curves to create a single curve with continuous curvature. This paper also presents an equation for finding the t value for each of the points selected to avoid overfitting and underfitting of the curve.