# TJU-Aerial-Robotics/YOPO

You Only Plan Once: A Learning Based Quadrotor Planner

Repository: https://github.com/TJU-Aerial-Robotics/YOPO
Canonical: https://ross.abutalabs.com/products/yopo
Language: C++
License: NOASSERTION
License Family: other
Last push: 2026-08-15T03:50:26+00:00

## Health v2 (maintenance only)
Score: 79/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 97, release rhythm 62, longevity 70
- inputs: {"age_days": 993, "days_push": 18, "days_rel": 253, "gap_med": 29, "n_releases_24m": 2}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1153, forks 144 (observed 2026-08-28T04:03:47.292571+00:00)

## What it is
YOPO is a learning-based one-stage planner for quadrotor autonomous navigation in obstacle-dense environments, integrating perception, mapping, path searching, and trajectory optimization into a single neural network. It uses motion primitive anchors with a guidance-learning training strategy that back-propagates trajectory cost gradients instead of imitation or reinforcement learning.

## Use cases
- plan trajectories for a quadrotor in obstacle-dense environments
- train a neural network drone planner with guidance learning
- build an autonomous navigation stack for a drone
- run agile tracking and navigation from perception to action
- replace classical front-end search and back-end optimization with a single network
- reproduce research on learning-based quadrotor planning

## When to choose
- you need fast, one-stage trajectory planning for quadrotors in cluttered spaces
- you want a learning-based planner without simulator-in-the-loop RL training
- you are doing research on drone navigation and want a strong open-source baseline
- you want matching open-source drone hardware designs for real-world experiments

## When to avoid
- you need a general-purpose planner for ground robots or manipulators
- you require a fully certified or safety-verified flight stack
- you want a plug-and-play commercial product rather than research code
- your project does not use ROS or C++

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, robotics, simulation
- domain: robotics, autonomous-vehicles, machine-learning, deep-learning
- platform: cpp
- tags: quadrotor, path-planning, drone-navigation, guidance-learning, motion-primitives, obstacle-avoidance, research-code, linux, ros

## Member repositories
- TJU-Aerial-Robotics/YOPO (main) score 79

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:47.292571+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:33:26.539020+00:00, confidence not recorded.
  - readme: https://github.com/TJU-Aerial-Robotics/YOPO (fetched 2026-08-28T04:03:47.292571+00:00, sha c4fd764fc6e7)
- Data as of 2026-08-30T08:39:29.467469+00:00.