Find Robot Operating System Projects

I'm the founder of a humanoid robotics AI company. I need a robotics/RL engineer who can work alongside me on real problems while explaining their thinking and approach so I build deep understanding over time.

This could lean more toward hands-on engineering or more toward structured teaching depending on your strengths. Ideally both.

Scope:

• Reinforcement learning for motor control (PPO, SAC, reward shaping, curriculum learning)
• Sim-to-real transfer pipelines for legged/humanoid robots
• Physics simulation environments (MuJoCo, Isaac Sim, Isaac Lab)
• Locomotion policy training and deployment
• Control systems fundamentals (PID, impedance control, EtherCAT)
• URDF/MJCF model creation and tuning
• Real-time inference at the edge

About me:

Strong software engineering background. I work daily with a robotics team and need to go from high-level AI/ML understanding to being able to evaluate technical decisions, read papers, and contribute meaningfully to architecture discussions.

Format:

• Video calls every 1-3 days (1-2 hours per session)
• Mix of collaborative engineering work and structured learning
• Async Q&A between sessions
• Duration: 2-3 months initially, ongoing if it works well

Ideal candidate:

• Has shipped sim-to-real on real legged or humanoid hardware
• Professional experience with MuJoCo and/or Isaac Sim
• Strong communicator who can explain complex concepts clearly
• Comfortable with a collaborative working style, not just isolated task execution

Budget: Open to hourly or weekly retainer. Share your rate and availability and we can arrange a short intro call to see if we are a fit.

I have a mobile robot prototype and I need an AI agent that will let it perceive its surroundings, decide on actions, and execute them safely in real time. Because the hardware stack is still flexible, I am open to ROS, a custom embedded setup, or a commercial SDK—whatever you feel will achieve the best balance between performance and maintainability.

At a minimum the agent must:
• Fuse camera and sensor data to detect obstacles and points of interest.
• Plan a collision-free path and update it on the fly.
• Expose a clean API so I can add higher-level skills later (object pickup, voice interaction, etc.).

Please structure the code in well-documented modules, supply build instructions, and include a short demo video that shows the robot handling a simple navigation scenario without remote intervention. Let me know the toolchain you prefer for simulation and testing so I can mirror the environment on my end.

Skills: Machine Learning (ML), Computer Vision, Robot Operating System (ROS), AI Text-to-speech, AI Chatbot Development, AI Model Development, AI Development, AI Agents

My name is Kaelan and I have had an interest in robotics for quite sometime. I have multiple concepts for robot designs, but I have very little technical expertise. I have a few projects related to robots and drones which I may be interested in pursuing depending on how complex the ideas are to accomplish. Ideally, I may be interested in finding a long-term business partnership with the right individual if they are interested in and see potential in my ideas.

Develop application software using ROS2 Jazzy, Nav2 and Raspberry Pi 4/5.

- take the existing software development environment (Docker, ROS2) https://github.com/makerspet/oomwoo-install
- implement in Gazebo simulation (and test thoroughly)
1. regular/auto cleaning (plan paths to cover the entire room/maps) https://github.com/makerspet/oomwoo/tree/main/contributions/cleaning-jobs
Starter tutorial https://makerspet.com/blog/simulate-oomwoo-one-robot-vacuum-in-gazebo-with-ros-2/ and here https://makerspet.com/blog/write-your-first-oomwoo-ros-2-package/
Achieve coverage 90%, time efficiency 80%
2. localize "kidnapped" robot https://github.com/makerspet/oomwoo/tree/main/contributions/nav-localize
Localize within 30 sec, 2 meters, 90% success rate.
3. Create regression test(s) for both on 1 world, run Gazebo headless using CLI
- Strip down existing ROS2/Docker development environment to product firmware running on Raspberry Pi 4 4GB. Drop Gazebo, desktop environment, etc. Make installation script. Boot, run on Pi 4. Use real placeholder vacuum, tutorial https://makerspet.com/blog/tutorial-connect-robot-vacuum-cleaner-to-ros-2-proscenic-m6-pro/
- 8 hours of community support and tutorials/documentation
- maintainer - handle PRs, talk to contributors, Apache 2.0 license, work in the open
- write step-by-step how-to tutorials for #1, #2, #3
- capture screen recording video of the simulation for the tutorials
Follow https://github.com/makerspet/oomwoo/blob/main/docs/SOFTWARE_INTERFACES.md guidelines.

Milestones
- M1 sim implementation + headless regression tests passing defined metrics
- M2 Pi 4 4GB image + install script, booting on real hardware
- M3 tutorials + docs + initial support

Out of scope - additional scope if you succeed

- optimize product firmware to run in 2GB RAM (instead of 4GB), including SLAM
- clean-and-map (vacuum doing mapping and cleaning simultaneously) high-level spec here https://github.com/makerspet/oomwoo/tree/main/contributions/clean-and-map
- implement recovery behaviors (to get robot unstuck)
- docking, undocking
- debug and test on real (but temporary placeholder) vacuum
- debug and test on actual vacuum
- make software production grade
- implement typical vacuum modes - spot cleaning, partition map into rooms, clean a room, clean large spaces (clean, charge, resume cleaning), vacuuming vs mopping, virtual wall support
- create additional worlds, regression tests
- more spec here https://github.com/makerspet/oomwoo/tree/main/contributions and here https://github.com/makerspet/oomwoo/blob/main/docs/RFC_BACKLOG.md
- implement cleaning blocked spaces (try slipping under couch covered by a cover that looks like obstacle)
- community support

Location: Remote
Type: Part-Time / Contract

Responsibilities
Develop and maintain applications using ROS/ROS2.
Integrate sensors, cameras, LiDAR, and robotic hardware.
Build and debug ROS nodes, topics, services, and actions.
Implement navigation, localization, and motion control.
Work with C++ and Python on Ubuntu/Linux.
Test and deploy robotic software on real hardware and in simulation.
Requirements
2+ years of hands-on experience with ROS/ROS2.
Strong proficiency in C++ and Python.
Experience with Linux, Git, and Gazebo/RViz.
Knowledge of SLAM, Navigation Stack/Nav2, TF, and sensor integration.
Experience working with mobile robots or robotic arms.
Ability to work independently and deliver on milestones.
Nice to Have
OpenCV or PCL
MoveIt
NVIDIA Jetson
Docker
AI/Computer Vision integration

To Apply: Share your resume along with GitHub, ROS projects, or videos demonstrating robots you've worked on.

i need a music box , please check the enclosed file , my requirement is in it ~
Please evaluate the R&D fee based on this simple architecture. Thanks!

I have a wheeled robot model inside CoppeliaSim and I need it to track a set of predefined waypoints using two complementary approaches: a classic PID controller and a Q-learning policy. Your job is to build both control loop of the python code for both pid and q learning

Skills: JavaScript, Python, Software Architecture, C++ Programming, Robotics, Lua, Data Analysis, Simulation, Robot Operating System (ROS), Reinforcement Learning

I have a mobile robot running inside CoppeliaSim and I need a clean, well-commented Python script that closes the loop on its position with a classic PID controller. The goal is simple: the robot must follow a predefined path accurately and smoothly, keeping tracking errors to a minimum while coping with CoppeliaSim’s real-time dynamics.

You are free to choose any sensible tuning method, but I want the gains justified and easy for me to tweak afterwards. Make sure the code exposes the proportional, integral and derivative terms clearly, logs position error over time, and plots or prints a brief performance summary so I can see how well the controller converges.

Deliverables:
• Python script (.py) ready to load in CoppeliaSim’s remote API or ZeroMQ interface
• Brief README or inline comments explaining setup, gain tuning procedure and how to change the reference path
• Proof-of-run: small video capture or screenshot series showing the robot completing the path within a reasonable tolerance (≈2 cm lateral error max)

Acceptance criteria:
– Robot completes the full path without oscillation or overshoot that causes it to leave the virtual floor area
– Average position error under 1 cm once steady tracking is reached
– All external libraries used must be standard Python or clearly listed for pip install

If you have tackled CoppeliaSim PID work before, feel free to mention it; otherwise, a concise demo will speak for itself.

Skills: PHP, JavaScript, Python, Software Architecture, Robotics, Data Visualization, Simulation, Robot Operating System (ROS)