Quintic Bézier curve and numerical optimal solution based path planning approach in seismic exploration

• Published in: Control engineering practice 2024-04, Vol.145, p.105855, Article 105855
• Main Authors: Niu, Tianwei, Wang, Liang, Xu, Yongkang, Wang, Junzheng, Wang, Shoukun
• Format: Article
• Language: English
• Subjects: Curvature constraint; Numerical optimization; Optimal path planning; Quintic Bézier curve
• ISSN: 0967-0661; 1873-6939
• DOI: 10.1016/j.conengprac.2024.105855

To guarantee the G2-continuity and the zero-curvature condition at the Vibroseis Source Points (VSPs), a novel optimal global path planning approach with operational constraints is proposed for vibroseis trucks in seismic exploration. This approach develops a simplified numerical optimization model based on parameterized quintic Bézier transition scheme, which allows for the quick and accurate solution of optimal path involving multiple constraints. Firstly, adopting pre-designed zero-curvature distribution scheme of control points, an elaborately parameterized quintic Bézier transition curve with a closed form is developed, which can satisfy the curvature constraints at the VSPs without redundant calculations. Secondly, considering the multiple objectives and constraints of the global path, the path planning problem is numerically modeled as an NLP problem to determine the optimal parameter set of Bézier transition curves. Then, to ensure fast convergence of the algorithm, the initial solution determination scheme is employed based on the geometric distribution of VSPs. Also, the boundary calculation is simplified by utilizing the convex hull property of the Bézier curve. Finally, the proposed path planning approach is verified by the EV56, a vibroseis truck applied in the seismic exploration, and the experimental results demonstrate that the average operation efficiency and precision can be increased by 6.83% and 9.61%, respectively. In field scenarios with unexpected obstacles, the vehicle successfully avoids the obstacles through local replanning. Meanwhile, the curvature profile and the safety margin outperformed the traditional method. [Display omitted] •Numerical optimization model based on parametric quintic Bézier transition scheme.•Satisfying G2-continuity and zero-curvature conditions through Bézier primitives.•Efficient collision detection based on convex hull property of Bézier curves.•Inequality constraints considering vehicle kinematics and abrupt curvature changes.•Efficient calculation of the cost gradient using closed-form parametric curve.