Technical Projects

Selected work in estimation, perception, learning, planning, and control for autonomous systems.

Visual Inertial SLAM
SLAM

Visual Inertial SLAM

Stereo-IMU-based VI-SLAM algorithm based on reference from ORB, OpenVINS, and VINSMono.

SLAM

Visual Inertial SLAM

Stereo-IMU-based VI-SLAM algorithm based on reference from ORB, OpenVINS, and VINSMono.

Period
November 2023

More Description about project in few days once implementation is done.

Q-Learning Controller
Reinforcement Learning

Q-Learning Controller

Pendulum swing-Up controller based on Q-Learning method.

Reinforcement Learning

Q-Learning Controller

Pendulum swing-Up controller based on Q-Learning method.

Period
November 2023

Developed a Q-learning controller for a multi-variate system, and tested on torque-controlled inverted pendulum. Improved learning efficiency by 73% & reduced learning time by 15% via integration of ϵ-greedy policy.

LQR/iLQR Controller
Reinforcement Learning

LQR/iLQR Controller

A LQR and iLQR controller for performing acrobatic action of 2D Quadrotor.

Reinforcement Learning

LQR/iLQR Controller

A LQR and iLQR controller for performing acrobatic action of 2D Quadrotor.

Period
November 2023

Developed & implemented an LQR and iLQR controller for precise position & trajectory tracking of a quadrotor in disturbance-prone environments with 0.01m accuracy. It is designed with line search optimization for acrobatic maneuvers & local minima avoidance

Semantic Segmentation
Deep Learning

Semantic Segmentation

Semantic segmentation to find drivable space using DeepLabV3 model.

Deep Learning

Semantic Segmentation

Semantic segmentation to find drivable space using DeepLabV3 model.

Period
November 2023

More Description about project in few days once implementation is done.

RNN-LSTM Trajectory Predictore
Neural Network

RNN-LSTM Trajectory Predictore

Trajectory prediction of surrounding traffic agents using RNN-LSTM algorithm.

Neural Network

RNN-LSTM Trajectory Predictore

Trajectory prediction of surrounding traffic agents using RNN-LSTM algorithm.

Period
November 2023

An RNN-LSTM Seq-2Seq model which was trained on the Apollo dataset for better autonomous motion. It is able to predict future 5 positions of surrounding traffic agents based on the past 10 position sequence. It was integrated with vehicle MPC control algorithm for smooth autonomous motion.

Kalman Filter Localization
Localization

Kalman Filter Localization

UAV Localization using EKF, UKF, and Optical Flow

Localization

Kalman Filter Localization

UAV Localization using EKF, UKF, and Optical Flow

Period
November 2023

Localize a quadrotor based on EKF and UKF by Sensor fusion of IMU and Vicon data. Developed an optical flow algorithm to localize the quadrotor and establish its velocity. Integrated visual Optical flow and IMU data to localize a quadrotor & compare with Vicon ground truth.