# 3we: AI-First Python API for Mobile Robot Navigation (Open Source, $300 BOM)

**URL:** <https://discourse.openrobotics.org/t/3we-ai-first-python-api-for-mobile-robot-navigation-open-source-300-bom/54872>\
**Category:** ROS General\
**Tags:** ros2, simulation, open-hardware, nav2, python\
**Created:** [May 15, 2026, 3:47pm UTC](https://discourse.openrobotics.org/t/3we-ai-first-python-api-for-mobile-robot-navigation-open-source-300-bom/54872 "2026-05-15T15:47:28Z")\
**Posts on this page:** 6\
**Page:** 1

<div class="post-metadata">

**Author:** ![telleroutlook](https://sea2.discourse-cdn.com/flex022/user_avatar/discourse.openrobotics.org/telleroutlook/32/33911_2.png) [@telleroutlook](https://discourse.openrobotics.org/u/telleroutlook)\
**Post date:** [May 15, 2026, 3:47pm UTC](https://discourse.openrobotics.org/t/3we-ai-first-python-api-for-mobile-robot-navigation-open-source-300-bom/54872/1 "2026-05-15T15:47:28Z")

</div>

Hi everyone,

I’d like to share \*\ ***3we** \*\* — an open-source platform I’ve been building that provides an AI-First Python API on top of ROS2/Nav2, targeting Embodied AI researchers who want to focus on algorithms rather than ROS2 infrastructure.

## The Problem

AI researchers (especially those working with VLMs/VLAs) often want to deploy models on real robots but face:

- Steep ROS2 learning curve (launch files, topics, services, actions)

- No clean path from simulation to hardware

- Existing platforms are either too expensive (TurtleBot 4: $1,200+) or simulation-only (Habitat, Isaac Lab)

## What 3we Does

```python

**from** threewe **import** Robot

**async**  **with** Robot(backend=“gazebo”) **as** robot:

```
image = robot.get_camera_image() # (H,W,3) uint8

scan = robot.get_lidar_scan() # LaserScan

**await** robot.move_to(x=5.0, y=3.0) # Nav2 under the hood

```

```

Change `backend=“gazebo”` to `backend=“real”` — same code runs on physical hardware. The ROS2/Nav2 stack is fully transparent to the user.

\*\ ***Four backends with identical API:** \*\*

- `mock` — zero-dependency 2D kinematics (no ROS2 needed, runs anywhere)

- `gazebo` — Gazebo Harmonic with full physics

- `isaac_sim` — NVIDIA Isaac Sim for GPU-accelerated RL training

- `real` — Physical hardware via ROS2 topics

## Architecture

```

┌─────────────────────────────────────────┐

│ AI-First Python API (user layer) │ ← Researchers write code here

├─────────────────────────────────────────┤

│ 3we-core (middleware) │ ← Backend dispatch, sensor fusion

├─────────────────────────────────────────┤

│ ROS2 / micro-ROS (infrastructure) │ ← Transparent to users

│ Nav2, slam\_toolbox, ESP32 drivers │

└─────────────────────────────────────────┘

```

This is NOT a replacement for ROS2 — it’s a layer on top that makes ROS2 accessible to ML researchers while preserving full ROS2 compatibility for roboticists who want low-level access.

## VLM-Controlled Navigation

The killer feature for AI researchers: GPT-4o (or any OpenAI-compatible VLM) can directly control the robot through natural language:

```python

**async**  **with** Robot(backend=“gazebo”) **as** robot:

```
result = **await** robot.execute_instruction(

    "find the red bottle and stop near it"

)

print(f"Success: {result.success}")

```

```

Internally this runs a perception-action loop: capture image → send to VLM → parse JSON action → execute → repeat until done. Works with GPT-4o, Qwen-VL, or local LLaVA.

## Hardware ($300 BOM)

Fully open reference hardware under CERN-OHL-P v2:

| Component | Selection |

|-----------|-----------|

| Compute | Raspberry Pi 5 (8GB) |

| AI Accelerator | Hailo-8L (13 TOPS) |

| MCU | ESP32-S3 + micro-ROS |

| LiDAR | LD06 (360°, 2D) |

| IMU | BNO055 (9-axis) |

| Drive | 4× N20 motors + Mecanum wheels + DRV8833 |

| Safety | Dual-channel relay (ISO 13850 E-stop) |

KiCad 8 PCB files, DXF mechanical drawings, and assembly docs all included.

## ROS2 Integration Details

For ROS2 developers who want to know what’s under the hood:

- \*\ ***Navigation** \*\*: Nav2 with DWB planner, parameters tuned for mecanum kinematics

- \*\ ***SLAM** \*\*: slam\_toolbox (online async) with LD06

- \*\ ***MCU bridge** \*\*: micro-ROS on ESP32-S3 via USB-C serial transport

- \*\ ***Sensor fusion** \*\*: robot\_localization EKF (IMU + wheel odometry)

- \*\ ***Launch** \*\*: Composable nodes, configurable via YAML profiles

You can always drop down to raw ROS2 topics/services if needed — the Python API doesn’t hide or lock you out.

## Benchmark Suite

7 standardized scenes with reproducible baselines:

```bash

threewe benchmark run --task pointnav --scene office\_v2 --episodes 100

```

Gymnasium-compatible environments for RL:

```python

**import** gymnasium **as** gym

env = gym.make(“3we/Navigation-v1”, scene=“office\_v2”, backend=“mock”)

```

## Demo

![Navigation Demo]([https://img.xuexiao.eu.org/1778836342754-19b737g.gif](https://img.xuexiao.eu.org/1778836342754-19b737g.gif))

Autonomous point-to-point navigation in office\_v2 scene with 360° LiDAR visualization.

## Links

- \*\ ***GitHub** \*\*: [GitHub - telleroutlook/3we-robot-platform: 3we Universal Modular Mobile Robot Platform — Open-source firmware, ROS2, hardware, and SDK · GitHub](https://github.com/telleroutlook/3we-robot-platform)

- \*\ ***Documentation** \*\*: [https://3we.org](https://3we.org)

- \*\ ***Paper** \*\*: [Paper - 3we](https://3we.org/paper)

- \*\ ***PyPI** \*\*: `pip install threewe`

Feedback welcome — especially from Nav2 users on whether the API abstraction makes sense, and from AI researchers on what’s missing for their workflows.

Software: Apache 2.0 | Hardware: CERN-OHL-P v2 | Docs: CC-BY-SA 4.0

---

<div class="post-metadata">

**Author:** ![chfritz](https://sea2.discourse-cdn.com/flex022/user_avatar/discourse.openrobotics.org/chfritz/32/26180_2.png) [@chfritz](https://discourse.openrobotics.org/u/chfritz)\
**Post date:** [May 15, 2026, 3:52pm UTC](https://discourse.openrobotics.org/t/3we-ai-first-python-api-for-mobile-robot-navigation-open-source-300-bom/54872/2 "2026-05-15T15:52:47Z")

</div>

Why do people keep posting unformatted markdown (copy and paste from AI it seems) and not bother fixing it?

---

<div class="post-metadata">

**Author:** ![Katherine\_Scott](https://sea2.discourse-cdn.com/flex022/user_avatar/discourse.openrobotics.org/katherine_scott/32/4434_2.png) [@Katherine\_Scott](https://discourse.openrobotics.org/u/Katherine_Scott)\
**Post date:** [May 15, 2026, 5:27pm UTC](https://discourse.openrobotics.org/t/3we-ai-first-python-api-for-mobile-robot-navigation-open-source-300-bom/54872/3 "2026-05-15T17:27:42Z")

</div>

I don’t know, but it is getting really annoying. I can tell they are AI generated because they all have some title like, “[Release] Some Title.” In this case I removed the unnecessary bracketed text. I don’t know why AI always inserts a bracketed tag in every title. We have built-in tags, please use them.

---

<div class="post-metadata">

**Author:** ![telleroutlook](https://sea2.discourse-cdn.com/flex022/user_avatar/discourse.openrobotics.org/telleroutlook/32/33911_2.png) [@telleroutlook](https://discourse.openrobotics.org/u/telleroutlook)\
**Post date:** [May 18, 2026, 11:51pm UTC](https://discourse.openrobotics.org/t/3we-ai-first-python-api-for-mobile-robot-navigation-open-source-300-bom/54872/4 "2026-05-18T23:51:00Z")

</div>

Apologies for the broken formatting — I pasted from my editor without checking the preview. I’ve reposted with proper formatting here: [Open Robotics in the Age of Embodied AI](https://discourse.openrobotics.org/t/open-robotics-in-the-age-of-embodied-ai/54933)  
Feel free to remove or close this original thread. Thanks for the heads-up.

---

<div class="post-metadata">

**Author:** ![Katherine\_Scott](https://sea2.discourse-cdn.com/flex022/user_avatar/discourse.openrobotics.org/katherine_scott/32/4434_2.png) [@Katherine\_Scott](https://discourse.openrobotics.org/u/Katherine_Scott)\
**Post date:** [May 19, 2026, 12:22am UTC](https://discourse.openrobotics.org/t/3we-ai-first-python-api-for-mobile-robot-navigation-open-source-300-bom/54872/6 "2026-05-19T00:22:22Z")

</div>

Please keep in mind that you are making a post that may be seen **by a few thousand Open Robotics Discourse users.** I suggest you approach composing a forum post with the same attention to detail that you would use for a keynote at a large conference. Just slapping the output of an LLM into the text box really isn’t a great idea.

Also keep in mind your audience, most of our members are either open source users or contributors. They are here to have meaningful conversations with other people working on different aspects of our projects. Most of them are not really interested in reading the output of your particular LLM session. I recommend using LLMs as an enhanced spell checker and not as the author of your posts.

---

<div class="post-metadata">

**Author:** ![Katherine\_Scott](https://sea2.discourse-cdn.com/flex022/user_avatar/discourse.openrobotics.org/katherine_scott/32/4434_2.png) [@Katherine\_Scott](https://discourse.openrobotics.org/u/Katherine_Scott)\
**Post date:** [May 19, 2026, 12:22am UTC](https://discourse.openrobotics.org/t/3we-ai-first-python-api-for-mobile-robot-navigation-open-source-300-bom/54872/7 "2026-05-19T00:22:30Z")

</div>