Quadratic optimization-based algorithm for driverless vehicle path smoothing
EDN: AIPWTJ
Abstract
Introduction (problem statement and relevance). When generating a driverless vehicle (DV) path or trajectory, the standard practice is to split the process in 2 stages – graph search and multi-objective optimization with obstacle avoidance. Before launching the second stage, the trajectory obtained through graph search needs to be smoothed preliminary in order to reduce contradiction between curvature significance and reference path following error significance.
The purpose of the study is to develop an algorithm for driverless vehicle path smoothing with possibility of intuitively clear adjustment of such parameters as deviation from reference path, path curvature and curvature change speed.
Methodology and research methods. To solve the problem, the method of analytical design is used for the path smoothing algorithm, which is based on using necessary function minimization conditions in order to synthesize the smoothing law.
Scientific novelty and results. An algorithm was developed for driverless vehicle path smoothing, which allows adjustment of deviation from the reference path, output path curvature and curvature change speed.
Practical significance. The proposed algorithm can be used to smooth the path obtained as a result of graph search for further use in path generation taking into account obstacle avoidance.
About the Authors
F. Yu. Belyakov
Federal State Unitary Enterprise “Central Scientific Research Automobile and Automotive Engines Institute” (FSUE “NAMI”)
Russian Federation
Russian Federation
Belyakov F.Yu. – postgraduate
Moscow 125438
V. V. Evgrafov
Federal State Unitary Enterprise “Central Scientific Research Automobile and Automotive Engines Institute” (FSUE “NAMI”)
Russian Federation
Russian Federation
Evgrafov V.V. – PhD (Phys-Math), director of the Center for intelligent systems
Moscow 125438
References
1. Kupreenko A.I., Isaev Kh.M., Mikhaylichenko S.M. [Autonomous driving systems]. [In innovations and technological breakthrough in the agro-industrial complex: proceedings of the International scientific and practical conference]. Bryansk, Bryanskiy gosudarstvennyy agrarnyy universitet Publ., 2020, part 2, pp. 249–254. EDN: BCTXNQ. (In Russian).
2. Scheuer A., Fraichard T. Planning continuous-curvature paths for car-like robots. IEEE/RSJ International Conference on Intelligent Robots and Systems. Osaka, Japan, 1996, рр. 1304–1311.
3. Buznikov S.E., Evgrafov V.V., Saykin A.M. Structuring tasks of control over driverless vehicles within intelligent transport systems. Journal of Physics: Conference Series. International Conference on Actual Issues of Mechanical Engineering, AIME, 2021, р. 012093. EDN: TDOPWF.
4. Endachev D.V., Bakhmutov S.V., Evgrafov V.V., Mezentsev N.P., Ryazantsev V.A., Khorychev A.A., Yusupov D.T. Implementation of modern approaches to development and testing of advanced driver assistance systems, including driverless systems. Journal of Physics: Conference Series. International Conference on Actual Issues of Mechanical Engineering, AIME, 2021, р. 012121. EDN: OONMZB.
5. Dubins L.E. On curves of minimal length with a constraint on average curvature, and with prescribed initial and terminal positions and tangents. American Journal of Mathematics, 1957, vol. 79, no. 3, pp. 497–516.
6. Evgrafov V.V. [System dynamics and control of whee led robots. Cand. phys. and math. sci. abstr. diss.]. Moscow, 2008. 15 p. Available at: https://new-disser.ru/_ vtoreferats/01003420648.pdf (accessed 14 August 2023). (In Russian)
7. Scheuer A., Fraichard T. Continuous-curvature path planning for car-like vehicles. IEEE International Confe rence on Intelligent Robots and Systems, 1997, рp. 997–1003.
8. Fraichard T., Scheuer A. From reeds and shepps to continuous-curvature paths. IEEE Transactions on Robotics, 2004, vol. 20, no. 6, pp. 1025–1035.
9. Bakolas E., Tsiotras P. On the generation of nearly optimal, planar paths of bounded curvature and bounded curvature gradient. American Control Conference, St. Louis, MO, 2009, pp. 385–390.
10. Piazzi A., Bianco C.G.L., Romano M. η³-splines for the smooth path generation of wheeled mobile robots. IEEE Transactions on Robotics, 2007, vol. 23 no. 5, pp. 1089–1095.
11. Modin K., Bogfjellmo G., Verdier O. A numerical algorithm for c²-splines on symmetric spaces. Available at: https://arxiv.org/pdf/1703.09589 (accessed 27 June 2023).
12. Stellato B., Banjac G., Goulart P., Bemporad A., Boyd S. OSQP: An operator splitting solver for quadratic programs. Mathematical Programming Computation, 2020, vol. 12, no. 4, pp. 637–672.
Review
For citations:
Belyakov F.Yu., Evgrafov V.V. Quadratic optimization-based algorithm for driverless vehicle path smoothing. Trudy NAMI. 2025;(3):54-60. (In Russ.) EDN: AIPWTJ
Views:
34
Email this article 
