Adaptive Navigation of a Transformer Robot in Warehouse Environments

Abstract

Developing an adaptive navigation system for transformer robots in warehouse environments is essential for improving efficiency and flexibility in operations. This paper proposes a novel adaptive navigation framework that integrates perception, path planning, robot reconfiguration, localization and path tracking into a unified solution. An overhead camera is used to capture the warehouse layout, and image processing methods are used to extract obstacles and free space through binary segmentation and contour detection. The processed environment is converted to a grid-based representation for path planning using a customized A* algorithm. This customized A* algorithm finds an efficient path for navigation considering the work envelope of the robot's shape and transformations with the available free space. For each path segment, it determines whether the robot should maintain its compact "O" configuration or transform into the elongated "I" configuration to access narrow passages. The localization of the robot in the environment is carried out through tracking an ArUco marker mounted on the robot enabling the path tracking capability during the navigation. The applicability of the proposed system has been validated through case studies conducted in scaled mock warehouse environments, demonstrating the potential of deploying transformer robots in warehouse environments.

Keywords

Cite this paper

@inproceedings{madhusanka2026adaptive,
  title     = {Adaptive Navigation of a Transformer Robot in Warehouse Environments},
  author    = {H.A.D. Madhusanka, R.M.S. Lakruwani, K.P.G. Sandamal, A.G. Tharindu Gimras, A.G.B.P. Jayasekara, M. A. Viraj J. Muthugala, Mohan R. Elara},
  booktitle = {ICIPRoB 2026 — IEEE},
  year      = {2026},
  publisher = {IEEE},
  doi       = {10.1109/ICIPRoB69625.2026.11497803},
  url       = {https://ieeexplore.ieee.org/document/11497803/},
}

Related Publications