yuwei-wu/RoboPhD

Some records and notes of weekly arXiv papers, GRASP seminars, and resources

RoboPhD

Some records of arXiv papers, GRASP seminars, and resources during my PhD period.

Workshops

Planning and UAVs

• ICRA 2024 Breaking Swarm Stereotypes
• ICRA 2024 Agile Robotics: From Perception to Dynamic Action
• ICRA 2024 Workshop on Field Robotics
• ICRA 2023 MW07: Energy Efficient Aerial Robotic Systems
• ICRA 2023 FW30: Active methods in autonomous navigation
• ICRA 2023 Bioinspired, Soft and Other Novel Design Paradigms for Aerial Robotics
• IROS 2023 Workshop on Integrated Perception, Planning, and Control for Physically and Contextually-Aware Robot Autonomy

Others

• CDC 2023 Workshop on Benchmarking, Reproducibility, and Open-Source Code in Controls
• IROS 2023 Workshop on Leveraging Models for Contact-Rich Manipulation

ArXiv Papers

Updates: I will not frequently update ArXiv papers on this repo, as Scholar Inbox is a much better tool for capturing papers!

Also, check robotics worldwide

The papers and notes are updated weekly, mainly about motion planning.

• 2021

• 2022-s1

• 2022-s2

• 2022-s3

• 2022-s4

• 2023-s1

• 2023-s2

Courses

• CS 4756/5756 Robot Learning, Cornell University

• CSCI 699: Robot Learning, USC

• 16-350 Planning Techniques for Robotics, CMU

• CSC2621 Topics in RoboticsReinforcement Learning in Robotics, UToronto

• CS 285 Deep Reinforcement Learning at UC Berkeley

• CS231n: Deep Learning for Computer Vision

• CIS 5150: Linear Algebra for Computer Vision, Robotics, and Machine Learning, Upenn

• ESE 546: Principles of Deep Learning, Upenn

• ESE 650 Learning in Robotics, Upenn

• CIS 7000: Large Language Models, Upenn

• MA 430 Differential Geometry

• CIS 610, Spring 2023 Advanced Geometric Methods in Computer Science, Upenn

• More Robotics Courses

Channels

GRASP Seminars

• 2022

• 2023

Others

• DeployableCoRL2023

Labs

These are the robotic labs I pay special attention to.

• MIT-ACL

• MIT-LIDS

• MIT-RLG

• MIT-LIS

• MIT-COCOSCI

• ZJU-FAST Lab

• Upenn-Kumar Lab

• TU Delft-Autonomous Multi-Robots Laboratory

• HKUST-Aerial Robotics Group

• UZH-RPG

• UCSD-ERL

• SNU-LARR

• UCB-HiPeR Lab

• CMU-Air Lab

• CMU-Robotic Exploration Lab

Resources

• Robotics

• Optimization

• Learning

• Computer Science

Guidelines

• Instructions to Ph.D. students by Prof.Dimitris Papadias
• Awesome tips for research
• Ten simple rules for structuring papers
• Novelty in Science: A guide to reviewers
• How to Write Mathmatics
• The Ten Most Important Rules of Writing Your Job Market Paper
• How to Write an Abstract
• How to Have a Bad Career in Research/Academia
• Doing a Systems PhD
• How to manage your time as a researcher
• A Normalized Professor Placement Guide to CS PhD Rankings
• A.I. Author Rankings by Publications
• Maximize your research impact with storytelling
• What advice would I give a starting graduate student interested in robot learning? Models! ... Model-free! ... Both!
• Science Research Writing: For Non-Native Speakers of English
• The Bitter Lesson

Faculty Application Resources

• The Strategy Space
• Tips for Computer Science Faculty Applications

Templates

• Review and Response Letters
• Best README
• Latex book

Knowledge Notes

• The Art of Linear Algebra
• Autonomous Racing Literature
• PhD Bibliography on Optimal Control, Reinforcement Learning and Motion Planning
• Deep Implicit Layers

Some Famous Planning/Control Repo

(1) Planner

• TEB Local Planner
• Fast Planner
• Teach-Repeat-Replan (Autonomous Drone Race)
• EGO-Planner-v2
• GPMP2
• MRSL Motion Primitive Library
• FASTER: Fast and Safe Trajectory Planner for Navigation in Unknown Environments
• cmu-exploration
• VAMP

(2) Multi-agent

• multi-robot-trajectory-planning
• Planner using Linear Safe Corridor
• MADER: Trajectory Planner in Multi-Agent and Dynamic Environments
• EGO-Swarm
• Downwash-Aware Trajectory Planning for Large Quadcopter Teams

(3) MPC /iLQR

• Model Predictive Contouring Controller (MPCC)
• Data-Driven MPC for Quadrotors
• Policy Search for Model Predictive Control with Application to Agile Drone Flight
• Model Predictive Control for Multi-MAV Collision Avoidance in Dynamic Environments
• MPC for Quadrotors with extension to Perception-Aware MPC
• KR iLQR Optimizer
• Online trajectory generation with distributed model predictive control for multi-robot motion planning

(4) Back-end Optimization

• Bilevel Planner
• DC3: A learning method for optimization with hard constraints
• GCOPTER

(5) Map representation

• Voxblox
• Grid Map
• OctoMap
• Convex Decomposition of Free Space

(6) Benchmarks

• Avoidbench
• Evaluating Dynamic Environment Difficulty for Collision Avoidance Benchmarking
• Design and Evaluation of Motion Planners for Quadrotors
• kinodynamic-motion-planning-benchmark
• Bench-MR: A Motion Planning Benchmark for Wheeled Mobile Robots
• Local Motion Planning Benchmark Suite

(7) Learning-based

• Fast Kinodynamic Planning on the Constraint Manifold with Deep Neural Networks
• k-diffusion

Robo Tools

(0) General tools

• Science Plots
• rosbag_fancy
• Manim, designed for creating explanatory math videos.
• Quick C++ Benchmark

(1) Solvers:

• ACADO Toolkit

  • Use: automatic control and dynamic optimization. It can solve MPC, but has some limits
  • License: open source
  • Interface: C++, with MATLAB

• CasADi

  • Use: nonlinear optimization and algorithmic differentiation
  • License: open source
  • Interface: C++, Python or Matlab/Octave

• FORCES PRO

  • Use: code generator for optimization solver, very useful to solve nonlinear MPC
  • License: Academic Licenses
  • Interface: C++, Python or Matlab /Simulink interface
  • Some examples: https://github.com/embotech/forcesnlp-examples

• Ceres Solver

  • Use: non-linear Least Squares with bounds constraints/ unconstrained optimization
  • License: open source
  • Interface: C++ library

• SeDuMi

  • Use: linear/quadratic/semidefinite solver
  • License: open source
  • Interface: Matlab/Octave

• Ensmallen

  • Use: non-linear numerical optimization
  • License: open source
  • Interface: C++ library

• HiGHS

  • Use: large scale sparse linear programming
  • License: open source
  • Interface: C, C#, FORTRAN, Julia and Python

• OR-Tools

  • Use: Google Optimization Tools
  • License: open source
  • Interface: C++, but also provide wrappers in Python, C# and Java

• CPLEX

  • Use: IBM optimization studio
  • License: have Free Edition

• Drake

  • Use: robotic toolbox, can solve optimizations, systems modeling, and etc.
  • License: open source
  • Interface: C++, python
  • Some examples: https://github.com/RobotLocomotion/drake-external-examples

• Crocoddyl

• YALMIP

• Mosek

  • Use: some types of optimizations. conic, QP, SDP...
  • License: Academic Licenses
  • Interface: C++, C, python, Matlab
  • Tutorials: https://github.com/MOSEK/Tutorials

• OOQP

  • Use: QP
  • License: MA27 from the HSL Archive
  • Interface: object-oriented C++ package

• Operator Splitting QP Solver (OSQP)

• Gurobi

  • Use: LP, QP and MIP (MILP, MIQP, and MIQCP)
  • License: Academic Licenses
  • Interface: C++, C, Python, Matlab, R...

• Embedded Conic Solver (ECOS)

  • Use: for convex second-order cone programs (SOCPs)
  • License: open source
  • Interface: C, Python, Julia, R, Matlab

(2) Simulations:

• Webots

• MuJoCo Physics

• Unreal Engine

Robotics Companies:

• Berkshire Grey
• fly4future