Robot platform

Drones

One flying platform for inspection, patrol and delivery

Hong Kong is a city of high-rise buildings, slopes and bridges, which makes it a natural place for drones that fly close to structures. A drone is a carrier platform: with the same flight controller and positioning system it can carry a camera, a LiDAR or a parcel box for different jobs.

Club teams start from a reliable platform, then choose an application, write a proposal and develop it. All flights follow the club safety rules and Hong Kong's rules for small unmanned aircraft, and take place only in approved areas with a mentor present.

Type
Robot platform
Applications
Inspection, patrol, monitoring, delivery
Key skills
Flight control, positioning near buildings, perception, flight safety
Who can lead
Tier 1 and Tier 2 members, by proposal

Applications

Each application is a possible project. Teams propose one and agree measurable targets with the review panel.

Building inspection

Close-range images of façades and curtain walls to find cracks, loose tiles and other damage, so that people do not have to work at height.

Infrastructure patrol

Routine inspection flights along slopes, bridges and construction sites, comparing images over time to spot changes.

Campus monitoring

Scheduled flights over the campus for mapping and event support, with automatic reports.

Delivery

Small parcel and medical supply delivery for Hong Kong's growing low-altitude economy.

Engineering foundations

The physics and positioning behind a reliable drone.

Rotor thrust

\[\begin{aligned} T_i &= k_f\,\omega_i^2 \\ \textstyle\sum_{i=1}^{4} T_i &= m\,g \ \text{(hover)} \end{aligned}\]

Each propeller gives thrust proportional to the square of its speed; together they must carry the drone and its payload.

Attitude dynamics

\[J\,\dot{\boldsymbol{\omega}} + \boldsymbol{\omega} \times J\boldsymbol{\omega} = \boldsymbol{\tau}\]

How the body turns under the torques from differences in rotor thrust. The flight controller closes this loop hundreds of times a second.

GNSS near tall buildings

\[\rho_i = \lVert \mathbf{p}_i - \mathbf{p} \rVert + c\,\delta t + \varepsilon_i\]

Reflections from buildings make the error ε large, so drones also fuse IMU, camera and LiDAR to keep an accurate position.

Photos

A delivery drone carrying a parcel
A delivery drone carrying a parcel.
Building inspection: flight paths close to a tower
Building inspection: flight paths close to a tower.
A delivery test flight by the harbour
A delivery test flight by the harbour.