SatAOI: Delimitating Area of Interest for Swing-Arm Troweling Robot for Construction

Jia‐Rui Lin, Shaobo Zhou, Peng Pan, Rong Cai, Gang Chen

Abstract

In concrete troweling for building construction, robots can significantly reduce workload and improve automation level. However, as a primary task of coverage path planning (CPP) for troweling, delimitating area of interest (AOI) in complex scenes is still challenging, especially for swing-arm robots with more complex working modes. Thus, this research proposes an algorithm to delimitate AOI for swing-arm troweling robot (SatAOI algorithm). By analyzing characteristics of the robot and obstacle maps, mathematical models and collision principles are established. On this basis, SatAOI algorithm achieves AOI delimitation by global search and collision detection. Experiments on different obstacle maps indicate that AOI can be effectively delimitated in scenes under different complexity, and the algorithm can fully consider the connectivity of obstacle maps. This research serves as a foundation for CPP algorithm and full process simulation of swing-arm troweling robots.

Introduction

Concrete troweling—the process of smoothing freshly poured concrete—is a critical, labor-intensive task in construction that traditionally demands high levels of manual skill. While the introduction of construction robots promises to automate this process, real-world construction sites present a significant challenge: they are unstructured environments filled with obstacles such as columns, reserved holes, and embedded parts.

To navigate these complexities, many modern troweling robots utilize a "swing-arm" design, consisting of a mobile tracked base and an independently swinging arm that holds the troweling disc. This design offers superior reach and flexibility but introduces complex kinematic constraints. Before a robot can execute a Coverage Path Planning (CPP) routine, it must first define its Area of Interest (AOI)—the specific regions of the floor the trowel can physically reach without any part of the robot colliding with obstacles. This paper introduces the SatAOI algorithm to solve this delimitation problem.

Methodology

The SatAOI algorithm follows a structured pipeline that moves from mathematical modeling to algorithmic search and final visualization.

Overview of the SatAOI framework, including mathematical modeling, search algorithms, and visualization of the Area of Interest (AOI).

1. Robot and Map Modeling

The environment is first discretized from CAD drawings into a high-resolution 10mm grid map. The robot itself is modeled as a three-part geometric system: a rectangular tracked vehicle (the base), a line-segment swinging arm, and a circular troweling disc. The robot's state is defined by four parameters: the base's coordinates ($x, y$), its rotation (yaw), and the swing-arm angle.

Mathematical model of the swing-arm troweling robot, defining the kinematics of the tracked base and the troweling disc.

2. Multi-Stage Collision Detection

To maintain computational efficiency, the algorithm employs a two-stage collision detection strategy. Initial checks use fast bounding volumes—a circumscribed circle for the base and a capsule shape for the arm—to quickly filter out distant obstacles. If an obstacle is detected within these bounds, precise geometric projection checks are performed. The algorithm utilizes KD-Trees to rapidly query the locations of nearby obstacles in the grid map.

The algorithm performs a Breadth-First Search (BFS) starting from a known safe point. The search is "trowel-centric," meaning it attempts to place the troweling disc in adjacent grid cells. For every potential cell, the algorithm tests 312 discrete angular combinations of the base and arm. If at least one configuration is found to be collision-free, that cell is marked as part of the AOI. This BFS approach ensures that the resulting AOI respects the physical connectivity of the site; the robot will not include areas it cannot physically reach due to impassable barriers.

Experimental Evaluation

The authors evaluated SatAOI using custom grid-based maps derived from real CAD floor plans of a gas-insulated switchgear (GIS) room. Three maps were created with increasing complexity:

  • Simple: 3.49% obstacle area.
  • Medium: 7.21% obstacle area.
  • Complex: 12.89% obstacle area, featuring obstacles ranging from 160mm to 2000mm.

Key Results and Insights

  • Robustness in Clutter: SatAOI successfully generated accurate AOIs across all map complexities. Even in the "Complex" scenario, the algorithm identified safe working zones while avoiding narrow gaps and corners where the robot's bulky base would be at risk.
  • Decoupled Kinematics: By treating the base and the arm/disc as separate entities with distinct bounding rules, the algorithm allows the robot to squeeze the trowel into tight spaces that a single, large bounding box would have marked as unreachable.
  • Efficiency: The combination of KD-Tree queries and a discrete angular resolution ($\pi/12$) allowed the algorithm to evaluate hundreds of poses per grid cell efficiently, making it viable for large-scale construction floor plans.

Conclusion

The SatAOI algorithm provides a robust foundation for the automation of concrete troweling. By accurately delimitating the reachable work area, it enables more effective Coverage Path Planning and full-process simulations. Future work will focus on integrating these AOI results into dynamic path-generation algorithms and exploring human-robot collaborative modes for extremely cluttered environments where full automation remains a challenge.

Summary updated

Abstract

In concrete troweling for building construction, robots can significantly reduce workload and improve automation level.However, as a primary task of coverage path planning (CPP) for troweling, delimitating area of interest (AOI) in complex scenes is still challenging, especially for swing-arm robots with more complex working modes.Thus, this research proposes an algorithm to delimitate AOI for swing-arm troweling robot (SatAOI algorithm).By analyzing characteristics of the robot and obstacle maps, mathematical models and collision principles are established.On this basis, SatAOI algorithm achieves AOI delimitation by global search and collision detection.Experiments on different obstacle maps indicate that AOI can be effectively delimitated in scenes under different complexity, and the algorithm can fully consider the connectivity of obstacle maps.This research serves as a foundation for CPP algorithm and full process simulation of swing-arm troweling robots.
BibTeX
@inproceedings{lin2025sataoi,
  author = {Jia‐Rui Lin and Shaobo Zhou and Peng Pan and Rong Cai and Gang Chen},
  title = {{SatAOI: Delimitating Area of Interest for Swing-Arm Troweling Robot for Construction}},
  booktitle = {Proceedings of the ... ISARC},
  year = {2025},
  doi = {10.22260/isarc2025/0052},
}
Figure