Terrain-adaptive motion planner for articulated construction vehicles in unstructured environments

• Published in: Automation in construction 2024-12, Vol.168, p.105864, Article 105864
• Main Authors: Huang, Tengchao, Chen, Xuanwei, Hu, Huosheng, Song, Shuang, Shao, Guifang, Zhu, Qingyuan
• Format: Article
• Language: English
• Subjects: Articulated construction vehicles; Motion planning; Stable motion trajectories; State estimation; Unstructured environments
• ISSN: 0926-5805
• DOI: 10.1016/j.autcon.2024.105864

In this paper, a terrain-adaptive motion planner is developed specifically for articulated construction vehicles (ACVs) to address instability issues caused by elevation changes on unstructured construction sites—challenges that traditional 2D motion planners struggle to manage effectively. The proposed planner adopts a modular framework, incorporating a terrain elevation model, an articulated vehicle kinematic model, and a posture response model. These models collaboratively capture the dynamic interactions between the vehicle and the terrain. The planner utilizes a multi-objective evaluation function to enhance the vehicle's 3D motion stability, especially in challenging terrains. By considering real-time vehicle-terrain interactions, this function estimates and optimizes the vehicle's stability. The planner's effectiveness is validated through field tests with a scaled-down ACV prototype, demonstrating significant improvements in stability and confirming its potential for practical application on unstructured terrains. •Developed a DWA-based motion planner to ensure stable operation of ACVs on unstructured terrains.•Extended trajectory representation to 3D space by coupling vehicle motion response with terrain excitation.•Optimized a multi-objective evaluation to balance different planning constraints.•Established a modular integrated architecture to support real-time updates.•Validated the planner's effectiveness through field tests with a scaled-down prototype.