Collision-free navigation of N-trailer vehicles with motion constraints

Leonardo Guevara, Miguel Torres-Torriti, Fernando Auat Cheein

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)


In this work, a collision-free navigation strategy for N-trailer vehicles is proposed. This approach is based on a scalable cascaded control scheme to perform several tasks simultaneously: trajectory tracking control, off-track error reduction, external obstacles avoidance, and inter-vehicle collision avoidance. To validate the proposed strategy, a Generalized N-trailer (GNT) structure with a car-like tractor and 10 trailers is tested in simulation to track an U-shape trajectory in presence of unknown obstacles, similar to the trajectories that agricultural vehicles must perform in real applications. The well-known information about external infrastructure is also considered to reduce unsafe trailers off-track errors in turning scenarios. Moreover, the motion constraints imposed by the car-like tractor physical limitations and the interconnections between trailers are also considered by restricting the control input in order to avoid collision between trailers. The simulation results obtained showed a safe navigation which performed feasible maneuvers without collisions between the vehicles' chain and any trailer or external obstacle.

Original languageEnglish
Title of host publication12th International Workshop on Robot Motion and Control 2019
Number of pages6
ISBN (Electronic)9781728129754
Publication statusPublished - 5 Aug 2019
Event12th International Workshop on Robot Motion and Control 2019 - Poznan, Poland
Duration: 8 Jul 201910 Jul 2019


Conference12th International Workshop on Robot Motion and Control 2019
Abbreviated titleRoMoCo 2019

ASJC Scopus subject areas

  • Artificial Intelligence
  • Mechanical Engineering
  • Control and Optimization


Dive into the research topics of 'Collision-free navigation of N-trailer vehicles with motion constraints'. Together they form a unique fingerprint.

Cite this