Author Archives: shihyunlo

Parametrized Propeller Model Online Learning

Propeller thrust models have been widely discussed in fluid mechanics and many existing models can incorporate its dynamic changes with relatively high accuracy. However, most of current platforms adopt simplified version assuming static airflow condition due to the complicated model formulation. This simplified

Parametrized Propeller Model Online Learning

Propeller thrust models have been widely discussed in fluid mechanics and many existing models can incorporate its dynamic changes with relatively high accuracy. However, most of current platforms adopt simplified version assuming static airflow condition due to the complicated model formulation. This simplified

Online Model Prediction and Control Adaptation

This work is based on the previous, “Propeller Model Online Learning”, and motivated to resolve the large modeling error particularly in situations while vehicles exploring in cluttered environments. Although the result of previous method seemed effective, it was not capable of

Online Model Prediction and Control Adaptation

This work is based on the previous, “Propeller Model Online Learning”, and motivated to resolve the large modeling error particularly in situations while vehicles exploring in cluttered environments. Although the result of previous method seemed effective, it was not capable of

Real-time SLAM for 3D Dense Mapping Using Geometric Features

The motivation of this project follows: Low cost 3D sensors (such as Kinect and Xtion) becomes popular and are applied on SLAM (Simultaneous localization and mapping) tasks. The large amount of 3D point cloud data produced by 3D sensors can

Real-time SLAM for 3D Dense Mapping Using Geometric Features

The motivation of this project follows: Low cost 3D sensors (such as Kinect and Xtion) becomes popular and are applied on SLAM (Simultaneous localization and mapping) tasks. The large amount of 3D point cloud data produced by 3D sensors can

Wind Field Prediction and Trajectory Optimization with Actuation Saturation

With the information of propeller efficiency from the actuation systems on quadrotors, we are capable of predicting the wind field around the vehicle. The fitted response surface below shows the prediction on wind velocity based on current motor states: Based

Wind Field Prediction and Trajectory Optimization with Actuation Saturation

With the information of propeller efficiency from the actuation systems on quadrotors, we are capable of predicting the wind field around the vehicle. The fitted response surface below shows the prediction on wind velocity based on current motor states: Based

Vision-aided State Estimator Study

This work is aimed to study the performance of using camera as the perception system for state estimator design on mobile robots. The system pipeline operates on ROS. Main study is focused on the robustness and accuracy of feature-based pose

Vision-aided State Estimator Study

This work is aimed to study the performance of using camera as the perception system for state estimator design on mobile robots. The system pipeline operates on ROS. Main study is focused on the robustness and accuracy of feature-based pose

Inspired by Twitch: Map Navigation with Crowdsourced Inputs

The power of crowdsourcing has been widely used since they can be easily accessed and the amount of data gives rise to a more complete and complicated set of models. In this project, it is focused on dealing with the problem that

Inspired by Twitch: Map Navigation with Crowdsourced Inputs

The power of crowdsourcing has been widely used since they can be easily accessed and the amount of data gives rise to a more complete and complicated set of models. In this project, it is focused on dealing with the problem that

Tree-Climbing Robot Design

The purpose of this research is to provide the service of coconut palm trimming on campus, to reduce the risk of the associated labor work. Therefore, the goal is to design a tree-climbing robot of self-reliance, capable of adapting to

Tree-Climbing Robot Design

The purpose of this research is to provide the service of coconut palm trimming on campus, to reduce the risk of the associated labor work. Therefore, the goal is to design a tree-climbing robot of self-reliance, capable of adapting to

Motion Planning with Uncertainties on Mobile Robot

In this assignment, a simple differential drive car needs to be controlled in a world filled with walls and open space. In addition to fixed walls, there will be road blocks which may or may not be open. The task is to execute a series

Motion Planning with Uncertainties on Mobile Robot

In this assignment, a simple differential drive car needs to be controlled in a world filled with walls and open space. In addition to fixed walls, there will be road blocks which may or may not be open. The task is to execute a series

Helicopter Controller: iLQR and Inverse Optimal Control

In this work, DDP was first used to find a controller that flies the helicopter following certain trajectory.  Inverse optimal control is then implemented based on the demonstrated trajectories from the given controller. The learned Q and R are computed

Helicopter Controller: iLQR and Inverse Optimal Control

In this work, DDP was first used to find a controller that flies the helicopter following certain trajectory.  Inverse optimal control is then implemented based on the demonstrated trajectories from the given controller. The learned Q and R are computed

Helicopter Hover Controller Learning

In this work, two methods are adopted and compared to learn a helicopter hover controller subjected to control input delay and noise in state estimate: neural network hover controller described in [Ng, Kim, Jordon, & Sastry, 2003], and the cross entropy

Helicopter Hover Controller Learning

In this work, two methods are adopted and compared to learn a helicopter hover controller subjected to control input delay and noise in state estimate: neural network hover controller described in [Ng, Kim, Jordon, & Sastry, 2003], and the cross entropy