Two-Stage Trajectory Optimization for Autonomous Ground Vehicles Parking Maneuver

Runqi Chai*, Antonios Tsourdos, Al Savvaris, Senchun Chai, Yuanqing Xia

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

67 Citations (Scopus)

Abstract

This paper proposes a two-stage optimization framework for generating the optimal parking motion trajectory of autonomous ground vehicles. The motivation for the use of this multilayer optimization strategy relies on its enhanced convergence ability and computational efficiency in terms of finding optimal solutions under the constrained environment. In the first optimization stage, the designed optimizer applies an improved particle swarm optimization technique to produce a near-optimal parking movement. Subsequently, the motion trajectory obtained from the first stage is used to start the second optimization stage, where gradient-based techniques are applied. The established methodology is tested to explore the optimal parking maneuver for a car-like autonomous vehicle with the consideration of irregularly parked obstacles. Simulation results were produced and comparative studies were conducted for different mission cases. The obtained results not only confirm the effectiveness but also reveal the enhanced performance of the proposed optimization framework.

Original languageEnglish
Article number8544047
Pages (from-to)3899-3909
Number of pages11
JournalIEEE Transactions on Industrial Informatics
Volume15
Issue number7
DOIs
Publication statusPublished - Jul 2019

Keywords

  • Autonomous ground vehicles
  • irregularly parked obstacles
  • optimal parking trajectory
  • particle swarm optimization (PSO)
  • two-stage optimization

Fingerprint

Dive into the research topics of 'Two-Stage Trajectory Optimization for Autonomous Ground Vehicles Parking Maneuver'. Together they form a unique fingerprint.

Cite this