Dr. Hao Peng's résumé


Dr. Hao Peng is now a Research Associate in the Mechanical and Aerospace Engineering Department at Rutgers University, NJ, USA. Hitherto I have published 45 academic journal and conference papers (see Google Scholar or my Publication List). I’m currently actively looking for a faculty position to continue my teaching and researching work.

Most recently, I’m exploring employing reinforcement learning techniques to handle some issues in deep space exploration.

I am working with Prof. Xiaoli Bai on several projects now. We are actively promoting the intersection of aerospace dynamics and machine learning techniques. Up to today, we have published a series of papers and presented several conference reports about using machine learning models to improve orbit prediction accuracy, which are validated using both simulated data and publicly available space object catalog data. We are still exploring the limits of this novel approach. Please see Machine Learning in Aerospace for details.

Before joining Prof. Bai’s research group at Rutgers, I received my PhD degree in Aerospace Engineering from Beijing University of Aeronautics and Astronautics (aka Beihang University). My PhD advisor was Prof. Shijie Xu who was a renowned professor in the aerospace area. My PhD thesis (in Chinese; available upon request) has been selected as Excellent Doctoral Degree Dissertation of the year 2016, among 10 theses from around 20 departments. Most of the content in my thesis have been published in English in top-ranking journals, please check Restricted Three-Body Problem below for details.

Short bio (2022-03-30 15:12:30)
Dr. Hao Peng has been a Research Associate at Rutgers University under the supervision of Prof. Xiaoli Bai. They are actively working on machine learning applications in aerospace engineering problems since 2016. Their exploration emphasizes on the uniqueness and exclusiveness of the problems that are suitable and may only be solved by the machine learning (ML) and artificial intelligence (AI) technology. For example, they have developed a machine learning approach to improve orbit prediction accuracy; a Bayesian optimization observation planning method for a constellation; a machine learning calibration method for MEMS gyro; etc. Dr. Peng’s research interests also include other general dynamics and operational problems.

Research Contributions

My detailed research contributions.

Research Interests

I have experience on various of topics:

  • General Aerospace Topics: Guidance, Navigation, and Control (GNC); Space Situational Awareness (SSA); Observation Scheduling;
  • Mechanics and Dynamics: Orbital mechanics; Restricted Three-Body Problem (RTBP) Model; Deep space exploration;
  • Trajectory Design & Optimization: Trajectory design and optimization; Low-thrust & low-energy transfer; Station-keeping strategy/control;
  • Computer Science: Machine learning; Data mining; Reinforcement learning; Visualization;

Research Experience

Curiosity and perfectionism are the driving force behind my career in scientific research.

Since 2011 when I starting my PhD studying (till the updating date of this page), I have published 15 papers and presents at top conferences every year. I really appreciate all the supports and instructions I have received from all my advisors and supervisors.

Most of my publications with my supervisor of the time were done through a collaborative working. The general direction was set up with the supervisor. Then, I was responsible for all the rest detailed works including implementations and analysis. I’m really enjoying this working flow because I can both learn the big vision of all my supervisors and practice organizing time and materials to accomplish a research project. These experience help me preparing for a future leading role, for example, either a PI (Principle Investigator) or a industrial manager.

Academic Publications and Services

My publication list.

My paper reviews.

Research Programming Skills

I use Git to manage all my research codes. My GitHub is here, but most of the work-related codes are managed in private repositories. BTW, my collection of useful third-party repositories therein is also worth a look.

Most of my works are accomplished using MATLAB.

Excellent skills.
Can write neat code.
Familiar with most toolboxes.
Experienced at visualizing.

Orekit is written in Java.
All simulations of orbital dynamics related to my machine learning publications (see Sec. Machine Learning in Aerospace) are carried out using Orekit, so I have been pretty experienced to using basic Java.
But I’m not capable of software development in Java yet.
Anyway, Java is a good language, easy to learn, having good rules to follow, and giving neat results.

Able to write simulations, modify third-party packages, visualize data.

Have experience to use Rattle in R to data mining.
We have published one paper using R.
R is a good and powerful language, especially for statistic studies, though it seems not quite popular in the astrodynamics/aerospace area.

Once “hacked” an R toolkit to transfer database from Mendeley to Zotero, which is not possible any more because Mendely encrypts their database to lock users in.

C is the first language I learned. C/C++ may be the ultimate language, but is a little bit hard to master.
I could read, use, and write simple codes, but never have a chance to write a big-scale simulations completely in C/C++.
Not much experience in pure C/C++ developing. Always have interests in hacking into GMAT.

I learned FORT77 in Barcelona by myself in just two weeks.
Modern fortran is no harder than FORT77, so I have confidence to start a new work in Fortran, if necessary.

In my spare time, I learned some Arduino tutorials.
Similar but much easier than C/C++.
A little bit complicated than MATLAB.

Learned through building my own website, to name a few: CSS, html, xml, json, JavaScript
I’m not an expert in any of these, but I can learn to make use of them if necessary.

Programming is just talking to machine.
Machine does not make error, but human does, because human makes machine.

More about Me

Find Me Online

There are plenty of ways to reach out to me:


Swimming, Soccer, Badminton, Ping-Pong,
Smartphones, Laptops,
Feeding animals in a zoo.