KingKongRobotics/jumper

选用指南

AI 根据 README 整理 · 2026/10/16 23:16:31

使用方式

在 AI 编程助手中打开此仓库，通过自然语言提示词进行外观设计、动作训练和场景创建。

查看 README 依据

 # Jumper **English** | [简体中文](README.zh.md) **Jumper is a 22-DoF crab robot.** [View hardware →](docs/HARDWARE.md) This repo is Jumper’s AI toolkit for appearance, motion, and scene creation. Open this repo in an AI coding assistant and try the prompts below. > 🦀 **Get a free Jumper!** [Find out how →](https://beunlimited.me/zh/events/crab-robot-challenge-2026) ## One sentence to design an appearance > Design a warm sand ranger appearance for Jumper with coordinated body and limb colors, then export a `.skin`. | | | | |:-:|:-:|:-:|

部署要求

训练基于 mjlab、rsl_rl、MuJoCo 和 MuJoCo Warp，需要 GPU 环境（如 Colab）。

查看 README 依据

| [Colab training](docs/COLAB.md) | GPU training in a notebook, checkpoint backup, resume, ONNX export and MP4 simulation replay. | | [Motion bundles](deploy/BUNDLE.md) | Package trained motions and their controller as an `.app`; see the [build guide](deploy/README.md) for prerequisites. | | [Project guide](docs/PROJECT_GUIDE.md) | Setup, current capabilities and further documentation. | Training builds on [mjlab](https://github.com/mujocolab/mjlab), [rsl_rl](https://github.com/leggedrobotics/rsl_rl), [MuJoCo](https://github.com/google-deepmind/mujoco)

已知限制

生成的输出不会自动继承本仓库的许可证。

查看 README 依据

and [MuJoCo Warp](https://github.com/google-deepmind/mujoco_warp). Example appearance and scene images come from jumper-design; [image sources](docs/media/DESIGN_SOURCES.md) and [third-party notices](NOTICE) record attribution. ## License Copyright 2026 KingKong Robotics. Maintainer-owned project materials are licensed under Apache-2.0. See [LICENSE](LICENSE), [NOTICE](NOTICE), and [licensing details](docs/PROJECT_GUIDE.md#license). Third-party materials remain under their respective terms, and generated outputs do not automatically inherit this repository's license.