©2018 by Yaru Niu. All rights reserved.

Projects

Whole-Body Imitation by Humanoid Robots and Task-Oriented Teleoperation

August 2016-May 2018

Human-Robot Intelligence Lab, SCUT

Advised by Prof. Zhijun Zhang

We developed a humanoid robot's whole-body imitation system enabling the imitation of head motions, arm motions, lower-limb motions, hand motions and locomotion and not requiring any ancillary handheld or wearable devices or any additional audio or gesture-based instructions. A task requiring standing on one foot, grasping and walking can be performed by teleoperation using this system in real time. As an extension to this system, we trained an offline imitation learning model using GMM and GMR (originally proposed by Calinon et al.).

The work including developing theory, designing algorithms and experiments, coding, writing and revising papers and patents, is mostly done by Yaru Niu, under the supervision of Prof. Zhijun Zhang.

Active Hierarchical Imitation and Reinforcement Learning in Continuous Tasks

Fall 2019

Georgia Tech CS 8803 - IRL

Advised by Prof. Matthew Gombolay

In this work, we developed an Active Hierarchical Imitation and Reinforcement Leaning (AHIRL) algorithm which can work in continuous spaces. We used DDPG to pre-trained the low-level controller, and employed DAgger to train our meta-controller from human demonstrations. What is more, two active learning methods were designed and used for initialize the agent's position in an episode.

 

Partners: Yijun Gu and Zuoxin Tang

Formation Control and Collision Avoidance using Multi-Agent Policy Gradient

Fall 2019

Georgia Tech ECE 6563 

In this project, we designed formation control and collision avoidance tasks with shaped reward functions, and explored two policy gradient methods, DDPG and MADDPG, which can be applied in the multi-agent environment.

Partners: Qian Luo and Sicong Jiang

DietMate -- A Multimodal Diet Monitoring System

This project is about monitoring people’s eating and drinking behaviors using the data obtained from multiple sensors, to help people maintain balanced diets and healthy life styles. In the project, I designed algorithms which employed signal processing techniques and machine learning algorithms to process the data, extracted features from the data and estimated the behaviors of eating and drinking.

Summer 2018

AICPS Lab, UC Irvine

Advised by Prof. Al Faruque

Analysis of Influencing Factors on Humanoid Robots' Emotion Expressions by Body Language

January 2018-April 2018

Human-Robot Intelligence Lab, SCUT

Advised by Prof. Zhijun Zhang

In this project, we explored humanoid robots’ capabilities of expressing emotions by body language and discussed the factors that may influence the emotion expression through questionnaire surveys and the statistic analysis. The research is carried out on the Nao robot and the results show that the expression of emotion is affected by the ambiguity of the body language and the joint limits of the robot.

Gesture - Determined Dynamical Schemes for Motion Planning of Humanoid Robot Arms

January 2018-February 2018 and May 2018

Zhike Intelligence Sci & Tech Co., Ltd.

Advised by Ziyi Yan and Prof. Zhijun Zhang

In this project, we explored and designed methods for Gesture - determined Dynamical Schemes for Motion Planning of the redundant manipulator, which can be applied to humanoid robots. In this way, the humanoid robot can not only perform end-effector tasks, but also can act following a sequence of specific gestures. Then we improved the robustness of the gesture-based dynamical scheme (avoided the joint limit) by modifying the gesture-based dynamical function which formulates the margins of the quadratic programming problem.

Partner: Lingdong Kong