一、Miniconda安装与环境创建
下载安装脚本：

wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-aarch64.sh

执行安装脚本：

bash Miniconda3-latest-Linux-aarch64.sh

创建新环境：

conda create -n yolov5 python=3.10

二、关键环境配置
进入环境：

conda activate yolov5

更新索引：

sudo apt-get update

安装基础依赖：

sudo apt-get install -y python3-pip
sudo apt-get install -y libjpeg-dev
sudo apt-get install -y zlib1g-dev
sudo apt-get install -y libpython3-dev
sudo apt-get install -y libopenblas-dev
sudo apt-get install -y libavcodec-dev
sudo apt-get install -y libavformat-dev
sudo apt-get install -y libswscale-dev 

安装重要依赖：

sudo apt install cuda-toolkit-12-6
pip install https://github.com/ultralytics/assets/releases/download/v0.0.0/torch-2.5.0a0+872d972e41.nv24.08-cp310-cp310-linux_aarch64.whl
pip install https://github.com/ultralytics/assets/releases/download/v0.0.0/torchvision-0.20.0a0+afc54f7-cp310-cp310-linux_aarch64.whl
wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/arm64/cuda-keyring_1.1-1_all.deb
sudo dpkg -i cuda-keyring_1.1-1_all.deb
sudo apt-get update
sudo apt-get -y install libcusparselt0
sudo apt-get -y install libcusparselt-dev

安装yolov5依赖：

pip install ultralytics
pip install opencv-python pillow matplotlib pandas seaborn scipy tqdm
pip install "numpy<2"

三、验证：
验证Jetpack版本：

sudo apt-cache show nvidia-jetpack

结果：
Source: nvidia-jetpack (6.2.2)
Version: 6.2.2+b24
Architecture: arm64
Maintainer: NVIDIA Corporation
Installed-Size: 194
Depends: nvidia-jetpack-runtime (= 6.2.2+b24), nvidia-jetpack-dev (= 6.2.2+b24)
Homepage: http://developer.nvidia.com/jetson
Priority: standard
Section: metapackages
Filename: pool/main/n/nvidia-jetpack/nvidia-jetpack_6.2.2+b24_arm64.deb
Size: 29308
SHA256: df018852245b906b4ac32e48704daa8024d1ba3624054eb563c28a7109d3e62b
SHA1: e82c20e59095ed5ef34793b95e1a46ec5e9f8d46
MD5sum: 47a7cece986656b89857d69b9ec0b79e
Description: NVIDIA Jetpack Meta Package
Description-md5: ad1462289bdbc54909ae109d1d32c0a8
验证CUDA版本：

nvcc --version

结果：
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2024 NVIDIA Corporation
Built on Tue_Oct_29_23:53:06_PDT_2024
Cuda compilation tools, release 12.6, V12.6.85
Build cuda_12.6.r12.6/compiler.35059454_0
验证GPU-pytorch环境：

import torch
print(torch.__version__)
print(torch.cuda.is_available())
if torch.cuda.is_available():
  print(f"当前 GPU 设备: {torch.cuda.get_device_name(0)}")

结果：
2.5.0a0+872d972e41.nv24.08
True
当前 GPU 设备: Orin

四、注意：
1.apt update后不用upgrade
2.pip install ultralytics会安装到~/miniconda3/envs/yolov5/lib/python3.10/site-packages，sudo pip install xxx会安装到/usr/lib/python3/...
3.