Pocket Queries and Konoike Construction Successfully Demonstrate Autonomous Exploration with Quadruped Robot in Mountain Tunnel Face
Pocket Queries and Konoike Construction successfully conducted a demonstration experiment for unmanned exploration using a quadruped robot at a mountain tunnel face. They successfully entered dangerous areas, performed gas monitoring, and acquired 3D point cloud data, proving the technology's effectiveness.
📋 Article Processing Timeline
- 📰 Published: May 26, 2026 at 00:10
- 🔍 Collected: May 25, 2026 at 15:32
- 🤖 AI Analyzed: May 25, 2026 at 15:53 (21 min after Collected)
Pocket Queries, Inc., in collaboration with Konoike Construction Co., Ltd., conducted a demonstration experiment on February 8, 2026, at the tunnel face (the forefront of excavation) of a mountain tunnel construction site. This experiment was part of the 'Construction Management Automation Project using Quadruped Robots' announced in July 2025.
This verification aimed to explore the possibility of unmanned exploration using quadruped robots in the 'tunnel face' working environment, which is the most dangerous area in mountain tunnel excavation. They successfully conducted unmanned exploration and data acquisition in a real-world setting, achieving a 'technical proof' that marks a major step toward practical application.
[Specific Results of the Demonstration Experiment]
In this experiment, the following three points were verified to confirm the acquisition of valid data:
1. Fully Unmanned Exploration of Dangerous Areas
Before human entry, a quadruped robot entered the tunnel face, which is prone to rock collapse immediately after blasting. It demonstrated that safe operation, through autonomous travel and remote control, is possible even on complex and uneven terrain.
2. Real-Time Hazardous Gas Monitoring
Equipped with gas detection sensors, the robot remotely monitored oxygen concentration and the presence of flammable/toxic gases near the tunnel face in real-time. This established a system to objectively determine environmental safety as a human substitute and quickly decide on entry feasibility.
3. High-Precision 3D Point Cloud Data Acquisition using LiDAR
Using LiDAR measurement, the robot acquired 3D point cloud data of the tunnel face and the tunnel structure. This data can be seamlessly linked with BIM/CIM models, contributing to the digitalization of construction management and the monitoring of changes over time.
[Future Developments]
Moving forward, the companies will focus on the following to improve robustness:
- Better adaptation to complex terrain: Improving walking stability on uneven ground.
- Establishing long-term operation: Achieving wide-range autonomous patrols through enhanced battery life and communication stability.
- Automating BIM/CIM integration: Implementing algorithms for automatic progress analysis.
The companies will continue to develop technologies that fundamentally transform safety and productivity at construction sites.
This verification aimed to explore the possibility of unmanned exploration using quadruped robots in the 'tunnel face' working environment, which is the most dangerous area in mountain tunnel excavation. They successfully conducted unmanned exploration and data acquisition in a real-world setting, achieving a 'technical proof' that marks a major step toward practical application.
[Specific Results of the Demonstration Experiment]
In this experiment, the following three points were verified to confirm the acquisition of valid data:
1. Fully Unmanned Exploration of Dangerous Areas
Before human entry, a quadruped robot entered the tunnel face, which is prone to rock collapse immediately after blasting. It demonstrated that safe operation, through autonomous travel and remote control, is possible even on complex and uneven terrain.
2. Real-Time Hazardous Gas Monitoring
Equipped with gas detection sensors, the robot remotely monitored oxygen concentration and the presence of flammable/toxic gases near the tunnel face in real-time. This established a system to objectively determine environmental safety as a human substitute and quickly decide on entry feasibility.
3. High-Precision 3D Point Cloud Data Acquisition using LiDAR
Using LiDAR measurement, the robot acquired 3D point cloud data of the tunnel face and the tunnel structure. This data can be seamlessly linked with BIM/CIM models, contributing to the digitalization of construction management and the monitoring of changes over time.
[Future Developments]
Moving forward, the companies will focus on the following to improve robustness:
- Better adaptation to complex terrain: Improving walking stability on uneven ground.
- Establishing long-term operation: Achieving wide-range autonomous patrols through enhanced battery life and communication stability.
- Automating BIM/CIM integration: Implementing algorithms for automatic progress analysis.
The companies will continue to develop technologies that fundamentally transform safety and productivity at construction sites.
FAQ
ポケット・クエリーズと鴻池組が行った実証実験の目的は何ですか？
山岳トンネル建設現場の切羽において、四足歩行ロボットを用いた無人探査の可能性を追求し、安全性と施工管理の自動化を検証することです。
実証実験でロボットが果たした役割は何ですか？
人間が立ち入る前の危険エリアへの先行進入、環境安全性を判断するためのリアルタイム有害ガスモニタリング、およびLiDARを用いた3D点群データの取得です。
使用されたロボットは何ですか？
Unitree B2-W という四足歩行ロボットです。
取得したデータはどのように活用されますか？
BIM/CIMモデルとシームレスに連携させ、施工管理のデジタル化や経時的な形状変化のモニタリングに活用されます。
今後のプロジェクトの課題は何ですか？
複雑地形への適応（歩行安定性向上）、長時間運用の確立（連続稼働時間および通信安定性の強化）、BIM/CIM連携の自動解析アルゴリズムの実装です。