B: Efficient and Optimal Base Placement for Fixed-Base Manipulators* was accepted by IEEE Robotics and Automation Letters (RA-L), 2025.

Authors: Zihang Zhao, Leiyao Cui, Sirui Xie, Saiyao Zhang, Zhi Han, Lecheng Ruan, and Yixin Zhu.

Links: Website · IEEE · ArXiv · Code

Overview

B* is an optimization framework for determining the optimal base placement of fixed-base manipulators across complete task trajectories. Instead of relying on precomputed reachability or inverse-kinematics databases, it works directly in configuration space and jointly reasons about base pose, robot joint configurations, task feasibility, joint limits, and collision constraints.

The method uses a two-layer strategy: an outer loop relaxes the fixed-base constraint and progressively tightens it toward one stable base pose, while an inner loop linearizes the non-convex kinematic constraints into tractable sequential linear programs. In large-scale evaluation, B* was tested across four manipulator types and 2,400 generated tasks, showing perfect success rates and substantially better placement optimality than sampling-based baselines.

Citation

1
2
3
4
5
6
7
8
9
10
@article{zhao2025b,
  title={B*: Efficient and Optimal Base Placement for Fixed-Base Manipulators},
  author={Zhao, Zihang and Cui, Leiyao and Xie, Sirui and Zhang, Saiyao and Han, Zhi and Ruan, Lecheng and Zhu, Yixin},
  journal={IEEE Robotics and Automation Letters},
  volume={10},
  number={10},
  pages={10634--10641},
  year={2025},
  doi={10.1109/LRA.2025.3604741}
}