In mine goafs, underground roadways, utility tunnels and tunnels, getting a drone to fly in and send back video is only the first step. For tasks that require verifying spatial structure, confirming coverage or pinpointing areas of interest, video alone is usually not enough. Video provides continuous visual evidence; a 3D point cloud turns the inspection flight into data that carries spatial coordinates. Only by combining the two can underground inspection move beyond a one-off viewing and become something that can be replayed, re-verified and delivered as a data product.
1. Video Shows You the Problem — It Does Not Always Tell You Where It Is
Live video is the baseline capability of underground inspection. Operators can observe roadways, walls, equipment, pipelines and obstacles on screen, and can record visible anomalies such as cracks, corrosion, deformation or foreign objects.
In goafs, mines and long utility tunnels, however, relying on video alone usually has several limits:
Footage is recorded in time order, so the spatial relationship between different locations is hard to reconstruct accurately;
When structures repeat and light is poor, it is very difficult afterwards to judge from video alone where exactly an anomaly is located;
Operators cannot easily see which areas have already been scanned and which may still be missed;
Results from different inspection rounds are hard to compare within one consistent spatial frame.
Video is therefore better at answering “what did we see on site”, while point-cloud data adds “where is it, what is the surrounding structure, and is the inspection coverage complete”.
2. A 3D Point Cloud Turns an Inspection Flight into Spatial Data
LiDAR continuously measures the distance to surrounding objects during flight and combines a large number of spatial sampling points into a point cloud. Combined with positioning and mapping methods such as SLAM, a drone can build a model of the surrounding space while moving, in an environment with no satellite positioning signal at all.
The value of a 3D point cloud is not that it replaces video with a more eye-catching picture. It adds spatial dimensions that can be used for verification and analysis:
Trajectory visualisation: the flight path and the spatial point cloud are displayed together, which helps make clear where the aircraft sits within the overall space;
Coverage checking: helps identify areas that have not yet been scanned or where data is sparse, reducing omissions caused by judging from memory;
Structural reconstruction: converts the geometry of roadways, chambers and equipment interiors into 3D data;
Marking points of interest: provides a spatial reference for re-checking anomalies, discussing risk and communicating with maintenance teams;
Deliverable archiving: produces data that can be reviewed, compared and post-processed over time.
3. Why Mine Goafs Need Point-Cloud Capability in Particular
A goaf typically combines no GPS signal, unknown geometry, insufficient light, many obstacles and a high risk for personnel entering. Conventional manual entry may be limited by collapse, standing water, dust or hazardous gas. A drone can enter first and provide remote imagery and spatial data, giving a reference for deciding whether people need to enter at all, which areas to inspect closely and how to plan remediation.
| Inspection need | Video | 3D point cloud |
|---|---|---|
| Real-time observation | Directly shows structures, obstacles and surface detail ahead | Helps make clear where the aircraft sits within the overall space |
| Coverage assessment | Depends on operator experience and video playback | Trajectory and point cloud are synchronised, helping identify areas not yet scanned |
| Structural reconstruction | 2D footage struggles to recover the complete structure | Generates a spatial point cloud for downstream 3D model processing |
| Later re-verification | Well suited to reviewing specific footage and anomaly detail | Well suited to locating, marking and re-checking within a consistent spatial frame |
4. How the Small Ball Huixun S300 Supports These Missions
The Small Ball Huixun S300 indoor anti-collision inspection drone from Aerofuture is designed for indoor and confined-space inspection. In GPS-denied or satellite-denied environments it can carry out image capture, spatial sensing, real-time point-cloud construction and 3D model generation. Its role is not simply to replace people, but to complete remote observation, spatial data acquisition and identification of key areas first, at times when personnel should not enter unknown or high-risk areas directly.
360° omnidirectional physical protection: the airframe uses a full-surround protective structure to reduce the risk of damage to the body and propellers from light contact in tight spaces;
Low-light image capture: a 4K visible-light camera, 1080P live image transmission and a 40-LED lighting array with a maximum brightness of 10,000 lumens;
LiDAR and SLAM positioning: a 905 nm LiDAR with 360° horizontal sensing and a maximum point-cloud output of 200,000 points per second, able to build a spatial point cloud in real time without GPS;
Synchronised trajectory and point cloud: the flight path and the spatial point cloud are presented together, helping operators observe the flight environment and identify areas that have not yet been scanned;
3D deliverable generation: inspection data can be imported into the companion system to generate a high-precision 3D model for structural review, locating key areas and subsequent analysis.
Small Ball Huixun S300 — key specifications
| Parameter | Specification |
|---|---|
| Dimensions | 365 × 352 × 200 mm (including protective cage) |
| Weight | 1,190 g (airframe) / 1,820 g (with battery) |
| Hover time | 15 min |
| Visible-light camera | 3840 × 2160, F1.8, 12.27 million effective pixels |
| Live image transmission | 1080P HD, latency as low as 80 ms |
| Lighting system | 40 high-efficiency LEDs, 10,000 lm |
| LiDAR | 905 nm, Class 1, 360° horizontal, up to 200,000 points per second |
| Positioning | Fused laser SLAM; hovering positioning accuracy below 2 cm |
| Maximum control distance | Up to 1,000 m; the effective distance in practice depends on site structure and communication conditions |
Application note: actual flight time, communication distance, positioning performance and point-cloud quality are affected by spatial structure, obstacles, target reflectivity, dust, temperature, communication obstruction, equipment configuration and operating practice. Specific performance and a suitable configuration must be assessed against the conditions on site.
5. Which Underground Tasks to Evaluate First
The Small Ball Huixun S300 can be used for mine goafs, underground roadways, utility tunnels, tunnels and other confined-space inspection. The main application scenarios include:
Mine goafs and underground roadways: what video can show, what a point cloud adds, and how it supports coverage verification;
Utility tunnels and pipe culverts: how to confirm coverage over long distances and how to assess communication obstruction;
Tunnels and air-raid shelters: how to position and map in GPS-denied, low-light environments with repetitive structures;
Boiler furnaces and equipment interiors: how high-intensity lighting supports observation of surface detail;
Ship cabins and enclosed spaces: how to reduce how often personnel must enter narrow, high-risk spaces directly;
Factories, warehouses and bridges: how to inspect ceilings, beams and columns and positions that are hard for people to reach.
6. What to Confirm Before a Project Starts
Entrance and passage dimensions, the tightest turning space, and whether there are trailing cables, protrusions or dense piping;
Whether the site is completely GPS-denied, and whether there is strong electromagnetic interference, metal obstruction, dust, standing water or high temperature;
Whether the objective is real-time reconnaissance, visible-light image capture or point-cloud modelling, or whether thermal imaging, gas detection or fibre-optic communication is also required;
The required deliverables, including video, flight trajectory, point-cloud files, 3D models or other project outputs;
Communication distance, wall and bend geometry, and whether a relay or another transmission method is needed;
Requirements for personnel safety, equipment protection, operator training and emergency response.
Configuration note: the visible-light camera is standard; a dual-light (thermal) lens corresponds to the S300T; gas sensors are customised; the fibre-optic module is optional and supports up to 1 km of wired fibre communication. Thermal imaging, gas detection and fibre-optic communication are selected according to the task, and the final configuration is subject to the project proposal.
7. Conclusion
Mine goaf inspection does not end with “we captured the footage”. Video solves on-site visualisation; a 3D point cloud solves spatial relationships, coverage verification and data retention. Only by combining live imagery, flight trajectory and point-cloud modelling can underground inspection gradually become a data chain that can be re-verified, analysed and used for later decisions.
Through omnidirectional protection, high-intensity lighting, HD imaging, LiDAR, real-time point cloud and 3D model generation, the Small Ball Huixun S300 can support preliminary observation and inspection data acquisition in confined spaces such as mines, utility tunnels, tunnels, boiler furnaces and ship cabins. Specific applications still require a technical assessment based on the site environment, mission objectives, equipment configuration and required data deliverables.
Frequently Asked Questions
Can video alone complete a mine goaf inspection?
Video can be used for real-time observation and recording. But if the project also needs to confirm inspection coverage, reconstruct the spatial structure, or produce data deliverables that can be re-verified later, flight trajectory and a 3D point cloud are normally required as well.
Can a 3D point cloud directly identify defects such as cracks and corrosion?
A 3D point cloud is mainly used to record spatial structure, positional relationships and scan coverage. Cracks, corrosion and temperature anomalies still need to be assessed together with HD imagery, thermal imaging, specialist inspection equipment or manual verification.
How does the drone position and map itself when there is no GPS signal?
The Small Ball Huixun S300 can combine LiDAR and SLAM technology to sense the surrounding structure and to position and map while flying. Actual performance is still affected by spatial structure, dust, obstacles and site conditions.
What data deliverables can the Small Ball Huixun S300 produce after an inspection?
Depending on project configuration and mission requirements, deliverables can include live video, flight trajectory, point-cloud files and 3D models. The exact outputs, formats and accuracy requirements should be confirmed with our technical team before the project starts.
What does the anti-collision structure do in a narrow space?
The omnidirectional protective structure reduces the risk of damage to the airframe and propellers if light contact occurs in a narrow space. A protective structure does not mean the drone can be flown into obstacles at will; it must still be operated by trained personnel in line with site conditions.
Are thermal imaging, gas detection and fibre-optic communication all standard?
No. The visible-light camera is standard; a dual-light lens corresponds to the S300T; gas sensors are customised according to project requirements; the fibre-optic module is optional. The final configuration should be confirmed against the actual mission.
Learn More About the Small Ball Huixun S300
For product documentation, equipment configuration and project fit for the Small Ball Huixun S300, see the Small Ball Huixun S300 product page or send us your inspection scenario. You can also call our national service hotline on 400-056-8989 or e-mail contact@aerofuture.cn.
