When drone research reaches real-flight validation, is the onboard computing power sufficient to run multiple perception workloads at the same time? When LiDAR, the gimbal, RTK, and the flight controller operate in parallel, is the data link real-time and reliable? In outdoor wind and large-area environments, is there enough margin in endurance, payload capacity, and flight stability?

As a project advances from single-UAV validation to multi-node, multi-sensor, and even multi-UAV collaboration, can the software architecture scale and remain maintainable over the long term?

The Prometheus 600 ROS 2 research drone development platform is now officially available.

This is not simply a matter of changing the ROS version on the P600. It combines a mature 600 mm large-wheelbase flight platform with a new generation of onboard computing and software designed for the ROS 2 ecosystem, integrating long-range electro-optical vision, 3D LiDAR perception, centimeter-level positioning, and autonomous planning in one platform.

From seeing the environment to understanding it, and from generating a path to executing a mission, the P600 ROS 2 provides a more complete starting point for developing complex autonomous flight capabilities.

01 

P600-ROS2

The P600 ROS 2 is a research drone platform designed for UAV research, algorithm validation, and engineering development. It uses an industrial-grade 600 mm large-wheelbase airframe and a Jetson Orin NX onboard computing platform with up to 157 TOPS of AI performance. It also integrates a GX40 electro-optical gimbal, a MID-360S 3D LiDAR, and dual-antenna high-precision RTK. The platform supports 3D mapping and localization, autonomous path planning, dynamic obstacle avoidance, object detection, and tracking. Its hardware, software, and development environment are already integrated, reducing equipment selection, environment deployment, and system integration work and helping developers move more quickly from simulation and algorithm development to real-flight validation.

How Should You Choose Between ROS 1 and ROS 2?

The ROS 1 and ROS 2 versions of the P600 use the same core hardware, including the airframe, flight controller, electro-optical gimbal, remote controller, and RTK. The main differences are in the onboard computing platform and software stack.

The newly released P600 ROS 2 uses:

Ubuntu 22.04 + ROS2 Humble + JetPack 6.2 + Prometheus V3

Prometheus V3 reorganizes system communication and functional modules around ROS 2 and introduces XRCE-DDS for bidirectional data transmission with the PX4 flight controller. Compared with a conventional architecture that depends on a Master node, ROS 2's decentralized communication model is better suited to parallel operation across multiple nodes and sensors. It also provides a clearer path for future multi-UAV collaboration, feature expansion, and long-term maintenance.

This means that the P600 ROS 2 has been systematically adapted to the ROS 2 ecosystem across flight-controller communication, functional modules, and the development toolchain.

 

Choose the ROS 1 version if you need to continue using an existing research group's ROS 1 packages. Choose the ROS 2 version for multi-UAV swarms, projects with real-time communication requirements, or projects that require long-term maintenance. For new robotics projects, continuously evolving enterprise R&D, and teams with higher requirements for communication latency and system scalability, this architecture makes it easier to advance from prototype validation to a stable engineering system.

02

Three Configurations for Different Development Needs

The P600 is available in three configurations covering 2D obstacle avoidance, 3D mapping, and ROS 2 development. For long-range target search, gimbal tracking, and aerial inspection, choose the GX40 + S3 configuration with 2D LiDAR obstacle avoidance. To run Fast-LIO SLAM, EGO-Swarm 3D path planning, and autonomous obstacle avoidance, choose the GX40 + MID-360S configuration with 3D LiDAR mapping and obstacle avoidance. Existing ROS 1 projects can use the ROS 1 version, while new projects or projects with higher computing and scalability requirements are recommended to use the 157 TOPS P600 ROS 2.

03

Computing Power Upgrade

100 TOPS Standard, Up to 157 TOPS in Super Mode

The P460 ROS 2 is equipped with the Allspark2 compact edge computer, powered by NVIDIA Jetson Orin NX. Standard mode provides 100 TOPS of AI performance, while Super mode delivers up to 157 TOPS. This gives the onboard system more GPU acceleration capacity for real-time tasks such as object detection and tracking, 3D SLAM, depth perception, and path planning.

Usability matters even more than the performance figure. Allspark 2 comes with the development environment preinstalled. Its compact size, low weight, and strong system portability reduce the preparation required for system flashing, dependency configuration, and functional verification. Together with the 410 mm modular airframe, composite structure, semi-enclosed protection, and optimized wiring, the platform achieves a practical balance between payload expansion and routine maintenance.

04

See Far, See All, and Locate Precisely: A Three-Layer Perception System

A drone platform built for complex missions must know where it is, understand what is around it, and lock onto distant targets.

The P600 ROS 2 combines the GX40 electro-optical gimbal, MID-360S 3D LiDAR, and dual-antenna high-precision RTK to form a three-layer perception chain spanning long-range vision, spatial perception, and precise positioning.

See All: MID-360S 3D LiDAR

The MID-360S provides a 360° × 59° field of view, a point rate of 200,000 points per second, and a near blind zone of 0.1 m. At 10% reflectivity, its detection range reaches 40 m. Its compact design and built-in IMU make it suitable for airborne deployment. Combined with Fast-LIO 3D LiDAR SLAM, the P450 ROS 2 can perform 3D mapping and localization while continuously supplying environmental data to EGO-Swarm path planning. Outdoors, the F9P positioning module can work with a ground base station and the P450 communication link to transmit RTCM data and achieve centimeter-level positioning.

See Far: GX40 Electro-Optical Gimbal

The GX40 uses a high-precision, three-axis non-orthogonal gimbal and an 8.29-megapixel camera. It supports 10x optical zoom and up to 40x combined zoom with digital magnification, and outputs 4K video at 30 fps.

Its built-in 850 nm laser illuminator provides an effective illumination range of up to 200 m for observation in low-light and completely dark environments. Combined with SpireCV-Pro, the platform supports object detection, click-to-track, gimbal lock-on, photography, and video recording, making it suitable for long-range target search, inspection, observation, and dynamic target tracking.

Locate Precisely: Dual-Antenna High-Precision RTK

The P600 ROS 2 is equipped with dual-antenna high-precision RTK405. Its built-in multi-constellation, multi-frequency positioning module and LoRa data link radio provide centimeter-level positioning without requiring RTCM transmission setup. The unit has its own battery and requires no external power supply. It is ready to use out of the box and easy to deploy.

It provides centimeter-level positioning and heading data for outdoor flight. A stable position reference improves hovering and trajectory control while providing a more reliable coordinate foundation for 3D mapping, long-range cruising, and multi-UAV missions.

05

600 mm Large Wheelbase for Greater Mission Margin

Running the algorithms is only part of the requirement; the airframe must also carry the load and fly steadily.

The P600 is built on the industrial-grade DP1000 flight platform. Its 600 mm large-wheelbase airframe and field-oriented control (FOC) propulsion system balance smooth motor operation, dynamic response, and flight efficiency. The aircraft supports Level 6 wind resistance and has a maximum takeoff weight of 4.2 kg. The P600 ROS 2 can hover in place for up to 29 minutes and 39 seconds. Longer flight time and greater payload margin support large-area mapping, outdoor inspection, target search, and other missions.

The included G16 integrated video and data link controller features a 7-inch sunlight-readable display for real-time viewing of flight status and video. With the complete aircraft configuration, the measured video and data transmission range is approximately 3.5 km, providing more convenient control and monitoring for outdoor tests and long-range missions.

More importantly, the P600 is not simply designed to be large. The electro-optical gimbal, 3D LiDAR, RTK, and onboard computer are integrated at the aircraft level, so developers do not have to solve mounting, power, communication, and software adaptation issues from scratch.

06

Core Functions

UAV Control Application

The control module in the Prometheus V3 software system enables precise UAV control. Acting as the bridge between application software and the aircraft, it provides multiple control states, including hover at the initial position, hover at the current position, landing, and movement. It supports position, velocity, and combined position-velocity control in both inertial and body frames, as well as trajectory control, attitude control, and latitude-longitude-altitude control (RTK). The module supports indoor and outdoor scenarios with MID360S or RTK positioning.

Object Detection

Load a YOLO object detection algorithm, such as YOLO11, to detect objects in the video stream. Operating instructions are provided so users can quickly replace the supplied model with their own YOLO model and accelerate algorithm validation.

Click-to-Track Target Tracking

Use a YOLO detection algorithm with a specified training dataset or the general-purpose COCO dataset to detect objects, then use tracking algorithms such as LiteTrack to track and lock onto the selected target. Outdoors, the gimbal can track people and other objects from general-purpose datasets, while the zoom gimbal enables ultra-long-range tracking.

FAST-LIO 3D SLAM Mapping

Using the MID360-S 3D LiDAR together with Fast-LIO, a tightly coupled LiDAR-inertial odometry algorithm, the system performs fast and robust 3D mapping and provides an environmental map for path planning and autonomous navigation. It supports switching between LiDAR-based and RTK positioning sources.

EGO-Swarm LiDAR Obstacle-Avoidance Path Planning

The platform supports the EGO-Swarm path planning algorithm, an upgrade to EGO-Planner, significantly improving autonomous operation in changing environments. The algorithm analyzes the surroundings in real time, avoids moving obstacles, and rapidly adapts to dynamically changing goal points, providing strong flexibility in unknown and complex scenarios. It is easy to use and works with the Pro ground station for autonomous 3D dynamic path planning. Obstacle-avoidance flight is supported at speeds of up to approximately 2 m/s.

07

One Toolchain Connecting Simulation, Perception, Ground Station, and Real Aircraft

Prometheus V3 Software System

Prometheus is an open-source autonomous UAV software platform that provides a complete solution for intelligent and autonomous flight. Building on the capabilities of V2, the Prometheus team developed Prometheus V3 with stronger support for the new-generation ROS 2 system, taking full advantage of ROS 2's real-time communication, system performance, and scalability. XRCE-DDS enables bidirectional data transmission with the PX4 flight controller using fewer system resources and greater communication efficiency. The release of Prometheus V3 marks a new stage in UAV development and provides enhanced solutions for mission planning, multi-UAV collaboration, intelligent perception, and other fields. Prometheus will continue to evolve, support more hardware platforms and emerging technologies, and accelerate the development of intelligent UAVs.

PrometheusSim Simulation Tool

PrometheusSim is a high-fidelity simulation tool for UAV development, research, and education. Built on the Unreal Engine, it renders detailed and realistic 3D scenes and supports simulation of lighting, shadows, weather, and complex terrain, providing a more immersive environment for algorithm validation, functional debugging, and simulation demonstrations.

Unlike a conventional Ubuntu + Gazebo simulation setup, PrometheusSim uses a collaborative Windows 11 + WSL2 architecture. High-quality simulation scenes run on Windows, while the PX4 flight controller and ROS packages run in WSL2, with close data exchange between the two environments. This balances graphics quality, development efficiency, and debugging convenience. A one-click installation script also greatly reduces complex environment configuration, allowing users to build a complete UAV simulation environment and begin learning, development, and testing more quickly.

SpireCV-PRO Intelligent Perception Library

SpireCV-Pro is an edge real-time perception development platform designed for intelligent unmanned systems and is an upgraded version of SpireCV. It provides extensive foundational capabilities across multiple platforms, including x86 PCs, Jetson, Rockchip, and Huawei Ascend. Its main functions include lightweight large-model inference, image- and LiDAR-based object detection and tracking, camera and gimbal control, video recording and streaming, and iterative edge data management. It is designed to provide mobile robotics developers with high-performance, reliable, feature-rich visual perception through concise interfaces.

PrometheusGroundStation-Pro System

PrometheusGroundStation-Pro is professional ground control software designed specifically for Prometheus UAV systems. The same installation supports both ROS 1 and ROS 2. Its efficient TCP/UDP communication mechanism greatly simplifies network configuration for multi-UAV collaboration and remote control, especially in swarm operations. The intuitive, easy-to-use interface runs across platforms and supports both real-flight and simulation environments, covering the complete workflow from algorithm validation to real-aircraft testing. Developers can also integrate custom functions quickly. The software currently covers core Prometheus capabilities including basic control, swarm control, gimbal control, visual tracking, and path planning, providing a convenient one-stop tool for UAV R&D and testing.

Resources

Wiki:

https://docs.amovlab.com/p600u-v2-wiki-p600ros2/template/preview/index.html#%E5%AF%BC%E8%AF%BB%2F%E5%AF%BC%E8%AF%BB.md