Helson Go is a Ph.D. Candidate with the FSC lab, specializing in slung payload motion planning and trajectory optimization.
Research Interests
Optimal control, slung payload control, multicopter flight dynamics, linear control
Education
Ph.D. – University of Toronto – Aerospace Engineering (2019-Present)
B.Eng – McGill University – Mechanical Engineering (2016 – 2019)
Publications
2023
Qin*, Chao; Yu*, Qiuyu; Go*, Shing Hei Helson; Liu, Hugh H. -T.
Perception-Aware Image-Based Visual Servoing of Aggressive Quadrotor UAVs Journal Article
In: IEEE/ASME Transactions on Mechatronics, 2023.
@article{TMech2023CQ,
title = {Perception-Aware Image-Based Visual Servoing of Aggressive Quadrotor UAVs},
author = {Chao Qin* and Qiuyu Yu* and Shing Hei Helson Go* and Hugh H. -T. Liu},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {IEEE/ASME Transactions on Mechatronics},
abstract = {The maintenance of visual features within the sensor field of view (FOV) poses a significant challenge for underactuated aerial vehicles like quadrotors, especially during aggressive maneuvers. However, existing image-based visual servo control (IBVS) methods rely on strict target visibility assumptions or impose excessive constraints on the quadrotor's agility to meet this requirement. Furthermore, the effectiveness of the visibility constraint defined in prior works remains unverified in aggressive flight tests. To address these issues, we present a robust IBVS scheme for quadrotors to perform aggressive maneuvers while ensuring target visibility. Based on the nonlinear model predictive control (NMPC) framework, we propose a novel underactuation compensation scheme to eliminate the need for a virtual camera frame, which enables us to formulate the true target visibility constraint. We then introduce a quaternion-based representation of spherical visual features to handle the nonsmooth vector field problem on the 2-sphere and derive its corresponding image kinematics. We validate our method through three challenging visual servo tasks where agile maneuvers are desired: fast landing, aggressive long-distance flight, and dynamic object tracking. Extensive simulation and experiment show that our method consistently achieves a target-visible rate of 100% in all image frames, even under a maximum pitch of 21.04$^circ$. The results validate the effectiveness of our visibility constraint under large robot rotations and underscore its importance in enabling robust and aggressive flights.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The maintenance of visual features within the sensor field of view (FOV) poses a significant challenge for underactuated aerial vehicles like quadrotors, especially during aggressive maneuvers. However, existing image-based visual servo control (IBVS) methods rely on strict target visibility assumptions or impose excessive constraints on the quadrotor's agility to meet this requirement. Furthermore, the effectiveness of the visibility constraint defined in prior works remains unverified in aggressive flight tests. To address these issues, we present a robust IBVS scheme for quadrotors to perform aggressive maneuvers while ensuring target visibility. Based on the nonlinear model predictive control (NMPC) framework, we propose a novel underactuation compensation scheme to eliminate the need for a virtual camera frame, which enables us to formulate the true target visibility constraint. We then introduce a quaternion-based representation of spherical visual features to handle the nonsmooth vector field problem on the 2-sphere and derive its corresponding image kinematics. We validate our method through three challenging visual servo tasks where agile maneuvers are desired: fast landing, aggressive long-distance flight, and dynamic object tracking. Extensive simulation and experiment show that our method consistently achieves a target-visible rate of 100% in all image frames, even under a maximum pitch of 21.04$^circ$. The results validate the effectiveness of our visibility constraint under large robot rotations and underscore its importance in enabling robust and aggressive flights.
Qin*, Chao; Yu*, Qiuyu; Go*, Shing Hei Helson; Liu, Hugh H. -T.
Perception-Aware Image-Based Visual Servoing of Aggressive Quadrotor UAVs Inproceedings
In: Advanced Intelligent Mechatronics (AIM), IEEE, 2023.
@inproceedings{AIM2023CQ,
title = {Perception-Aware Image-Based Visual Servoing of Aggressive Quadrotor UAVs},
author = {Chao Qin* and Qiuyu Yu* and Shing Hei Helson Go* and Hugh H. -T. Liu},
url = {https://www.youtube.com/watch?v=X2-SMGD99oA},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {Advanced Intelligent Mechatronics (AIM)},
journal = {IEEE/ASME International Conference on Advanced Intelligent Mechatronics},
publisher = {IEEE},
abstract = {The maintenance of visual features within the sensor field of view (FOV) poses a significant challenge for underactuated aerial vehicles like quadrotors, especially during aggressive maneuvers. However, existing image-based visual servo control (IBVS) methods rely on strict target visibility assumptions or impose excessive constraints on the quadrotor's agility to meet this requirement. Furthermore, the effectiveness of the visibility constraint defined in prior works remains unverified in aggressive flight tests. To address these issues, we present a robust IBVS scheme for quadrotors to perform aggressive maneuvers while ensuring target visibility. Based on the nonlinear model predictive control (NMPC) framework, we propose a novel underactuation compensation scheme to eliminate the need for a virtual camera frame, which enables us to formulate the true target visibility constraint. We then introduce a quaternion-based representation of spherical visual features to handle the nonsmooth vector field problem on the 2-sphere and derive its corresponding image kinematics. We validate our method through three challenging visual servo tasks where agile maneuvers are desired: fast landing, aggressive long-distance flight, and dynamic object tracking. Extensive simulation and experiment show that our method consistently achieves a target-visible rate of 100% in all image frames, even under a maximum pitch of 21.04$^circ$. The results validate the effectiveness of our visibility constraint under large robot rotations and underscore its importance in enabling robust and aggressive flights.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The maintenance of visual features within the sensor field of view (FOV) poses a significant challenge for underactuated aerial vehicles like quadrotors, especially during aggressive maneuvers. However, existing image-based visual servo control (IBVS) methods rely on strict target visibility assumptions or impose excessive constraints on the quadrotor's agility to meet this requirement. Furthermore, the effectiveness of the visibility constraint defined in prior works remains unverified in aggressive flight tests. To address these issues, we present a robust IBVS scheme for quadrotors to perform aggressive maneuvers while ensuring target visibility. Based on the nonlinear model predictive control (NMPC) framework, we propose a novel underactuation compensation scheme to eliminate the need for a virtual camera frame, which enables us to formulate the true target visibility constraint. We then introduce a quaternion-based representation of spherical visual features to handle the nonsmooth vector field problem on the 2-sphere and derive its corresponding image kinematics. We validate our method through three challenging visual servo tasks where agile maneuvers are desired: fast landing, aggressive long-distance flight, and dynamic object tracking. Extensive simulation and experiment show that our method consistently achieves a target-visible rate of 100% in all image frames, even under a maximum pitch of 21.04$^circ$. The results validate the effectiveness of our visibility constraint under large robot rotations and underscore its importance in enabling robust and aggressive flights.
Go*, Shing Hei Helson; Qian*, Longhao; Liu, Hugh HT
Data-Driven and Robust Path-following Control of a Quadrotor Slung Load Transport System Inproceedings
In: AIAA SciTech, 2023.
@inproceedings{SciTech2023HG,
title = {Data-Driven and Robust Path-following Control of a Quadrotor Slung Load Transport System},
author = {Shing Hei Helson Go* and Longhao Qian* and Hugh HT Liu},
url = {https://www.flight.utias.utoronto.ca/fsc/wp-content/uploads/2023/05/scitech2023hg_final.pdf},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {AIAA SciTech},
journal = {AIAA SciTech},
abstract = {In this paper, a robust path following control law for a quadrotor Slung Load Transport System is developed. A Gaussian Process-augmented Extended Kalman Filter is proposed to estimate payload states. In this approach, Gaussian Processes are used to compensate for unmodelled dynamics in the process model, and they are trained on previously collected data of a Slung Load Transport System in flight. Both simulations and experiments verify the estimation and control system framework and demonstrate successful stabilization and trajectory tracking of the Slung Load Transport System, overcoming model inaccuracy and disturbances. I. Nomenclature F − → = NED inertial frame F − → = Quadrotor body-fixed frame, fixed at the mass center of the quadrotor vehicle = Quadrotor mass J = Quadrotor moment of inertia = Payload mass = Cable length g = Gravity vector in the world frame ∈ R 3 = Vector spanning the length of the cable x ∈ R 3 = Absolute position of the payload mass center v ∈ R 3 = Absolute velocity of the payload mass center, expressed in F − → r ∈ R 2 = Relative position of the quadrotor mass center w.r.t. the payload, projected onto the XY plane of F − → v ∈ R 2 = Relative velocity of the quadrotor mass center w.r.t. the payload, projected onto the XY plane of F − → ˜ v ∈ R 2 = Difference between v predicted by some dynamics model and the true value B ∈ R 3×2 = Projection from v to the time derivative of R ∈ SO(3) = Rotation of F − → relative to F − → ∈ R 3 = Angular velocity of the F − → relative to F − → , expressed in F − → f ∈ R 3 = Total thrust delivered by the actuators ∈ R 3 = Total torque delivered by the actuators d ∈ R 3 = Disturbance acting on the payload d ∈ R 3 = Disturbance acting on the quadrotor body d ∈ R 3 = Sum of disturbances acting on the SLTS e ∈ R 3 = Positional (radial) error of the payload from some given path e ∈ R 3 = Velocity (tangential) error of the payload from some given path × : R 3 → R 3×3 = Maps a 3-vector to a 3-by-3 skew-symmetric (cross product) matrix ∨ : R 3×3 → R 3 = The inverse operation of the × mapping * Ph.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
In this paper, a robust path following control law for a quadrotor Slung Load Transport System is developed. A Gaussian Process-augmented Extended Kalman Filter is proposed to estimate payload states. In this approach, Gaussian Processes are used to compensate for unmodelled dynamics in the process model, and they are trained on previously collected data of a Slung Load Transport System in flight. Both simulations and experiments verify the estimation and control system framework and demonstrate successful stabilization and trajectory tracking of the Slung Load Transport System, overcoming model inaccuracy and disturbances. I. Nomenclature F − → = NED inertial frame F − → = Quadrotor body-fixed frame, fixed at the mass center of the quadrotor vehicle = Quadrotor mass J = Quadrotor moment of inertia = Payload mass = Cable length g = Gravity vector in the world frame ∈ R 3 = Vector spanning the length of the cable x ∈ R 3 = Absolute position of the payload mass center v ∈ R 3 = Absolute velocity of the payload mass center, expressed in F − → r ∈ R 2 = Relative position of the quadrotor mass center w.r.t. the payload, projected onto the XY plane of F − → v ∈ R 2 = Relative velocity of the quadrotor mass center w.r.t. the payload, projected onto the XY plane of F − → ˜ v ∈ R 2 = Difference between v predicted by some dynamics model and the true value B ∈ R 3×2 = Projection from v to the time derivative of R ∈ SO(3) = Rotation of F − → relative to F − → ∈ R 3 = Angular velocity of the F − → relative to F − → , expressed in F − → f ∈ R 3 = Total thrust delivered by the actuators ∈ R 3 = Total torque delivered by the actuators d ∈ R 3 = Disturbance acting on the payload d ∈ R 3 = Disturbance acting on the quadrotor body d ∈ R 3 = Sum of disturbances acting on the SLTS e ∈ R 3 = Positional (radial) error of the payload from some given path e ∈ R 3 = Velocity (tangential) error of the payload from some given path × : R 3 → R 3×3 = Maps a 3-vector to a 3-by-3 skew-symmetric (cross product) matrix ∨ : R 3×3 → R 3 = The inverse operation of the × mapping * Ph.
1 thought on “H S Helson Go”
Comments are closed.