About me

> Download Resume PDF

I am a master's student focusing on reinforcement learning, deep learning and machine learning.
I developed projects relating to robot navigation based on deep reinforcement learning, in which a paper was accepted by ICRA2022. Over the past years, I also worked in several big companies, where I developed several Android apps, Java backend programs, and vue.js frontend programs.


* Education *

2019 - 2022     Msc, University of Bonn      Computer Science; English; Germany

2014 - 2018     Bas, Shandong University     Software Engineering; Bilingual; China


* Work *

2021 - 2022     Humanoid Robots Lab      Student Research Assistant; Germany

2020 - 2022     Fraunhofer IAIS          Software Engineer; Working Student; Germany

2017 - 2022     Zhuhai Guangge Technology      Software Engineer; Remote; China

2017 - 2017     JD.com              Java Developer Intern; China


* Publication *

Subgoal-driven robot navigation in indoor environments

Weixian Shi

Humanoid Robots Lab

Master Thesis

code / video


Enhanced Spatial Attention Graph for Motion Planning in Crowded, Partially Observable Environments

Weixian Shi, Yanying Zhou, Xiangyu Zeng, Shijie Li, Maren Bennewitz

Humanoid Robots Lab

ICRA 2022

paper / code / video


* Project *

Mechanisms of Action (MoA) Prediction

Deep Learning competition (Python, Pytorch, Deep Learning)

Kaggle.com

Integrate data from multiple MySQL databases, find and aggregate the relationship between major shareholders and companies in massive data, and store the results in Redis. In this project, I learned how to quickly and efficiently aggregate useful information from large amounts of data.

code


Home page of Fraunhofer BIG DATA AI

Frontend program (Html, Vue.js)

Fraunhofer IAIS

link

Reconstruct the entire page of the Fraunhofer Big Data AI homepage, connecting data from Google spreadsheets and presenting the data by category. At the same time, the project list for each location is summarized and displayed on Google Maps.


Investment relation map for JD stock app

Java backend program (Java, Mysql, Redis)

JD.com

Integrate data from multiple MySQL databases, find and aggregate the relationship between major shareholders and companies in massive data, and store the results in Redis. In this project, I learned how to quickly and efficiently aggregate useful information from large amounts of data.


A realtime web-based match system for electronic laser shooting sports events

Java backend program (Java, Mysql, Vue.js, Websocket)

Zhuhai Guangge Technology Co., Ltd.

An online competition system based on Springboot, MyBatics, vue.js, WebSocket, etc., synchronizes the shooting scores of the Android app to the Java background in real-time and updates the ranking of the players. The system integrates WebSocket to obtain the synchronization of real-time data and instructions between Android, Java backend, and frontend. At the same time, Redis is used to cache the ranking information of the players and speed up the access speed.


Android APP of laser shooting systems for sporting events

Android (Java, Bluetooth, Visualization, Websocket)

Zhuhai Guangge Technology Co., Ltd.

An app for middle school student-athletes that receives hardware device data via Bluetooth, and visualizes laser transmitter firing positions, trigger pressure data, and more. This app can receive and parse data from different types of devices. In addition, synchronous backup of data is also implemented based on socket and http, as well as its use in competition scenarios.


Official website of Zhuhai Guangge Technology Co., Ltd.

Frontend program (html, css)

Zhuhai Guangge Technology Co., Ltd.

link

A static resource website and deployed on nginx



© 2022 weixians.github.io