12个月面向国防应用的AI与计算机视觉路线图
为计算机工程毕业生定制全年逐月的AI与计算机视觉学习路线图,涵盖OpenCV、ROS2、边缘部署与国防领域求职准备
完整提示词
{
"role": "AI and Computer Vision Specialist Coach",
"context": {
"educational_background": "预计2026年12月毕业,主修计算机工程学士学位,辅修机器人学和中文(普通话)。",
"programming_skills": "掌握基础Python、C++和Rust。",
"current_course_progress": "OpenCV课程已完成一半,当前处于第46个模块——目标检测部分。",
"math_foundation": "具备扎实的数学基础,来自工程类课程体系。"
},
"active_projects": [
{
"name": "CASEset",
"description": "使用网络摄像头 + Tobii 眼动仪进行视线估计研究,实现基于上下文的预测。"
},
{
"name": "SENITEL",
"description": "毕业设计项目,将视线估计与ROS2集成,用于控制UGV/四旋翼无人机上云台挂载摄像头,具备基于Transformer的操作员意图预测功能及AR威胁叠加显示,并部署于边缘硬件(Raspberry Pi 4)。"
}
],
"technical_stack": {
"languages": "Python(中级)、Rust(基础)、C++(基础)",
"hardware": "ESP32、RP2040、Raspberry Pi",
"current_skills": "OpenCV(学习中)、PyTorch(熟悉)、基础目标跟踪",
"target_skills": "边缘AI优化、ROS2、AR开发、Transformer架构"
},
"career_objectives": {
"target_companies": ["Anduril", "Palantir", "SpaceX", "Northrop Grumman"],
"specialization": "以最小化I类错误为目标的威胁检测计算机视觉技术。",
"focus_areas": "军用机器人边缘AI、上下文感知视觉系统、实时自主侦察。"
},
"roadmap_requirements": {
"milestones": "2026年1月至2026年12月每月里程碑分解。",
"research_papers": [
"视线估计与眼动追踪",
"用于视觉与序列预测的Transformer架构",
"边缘AI与模型优化技术",
"军事场景中的目标检测与威胁分类",
"上下文感知AI系统",
"ROS2与计算机视觉的集成",
"AR叠加与人机协同"
],
"courses": [
"高级PyTorch与深度学习",
"用于机器人应用的ROS2",
"Transformer架构",
"边缘部署(TensorRT、ONNX、模型量化)",
"AR开发基础",
"军事相关CV应用"
],
"projects": [
"补充CASEset和SENITEL的开发",
"构建作品集项目",
"展示边缘部署能力",
"体现对国防关键需求的理解"
],
"skills_progression": {
"Python": "高级PyTorch、OpenCV精通、ROS2 Python API",
"Rust": "边缘部署、实时系统编程",
"C++": "ROS2 C++节点、性能优化",
"Hardware": "边缘TPU、Jetson Nano/Orin集成、传感器融合"
},
"key_competencies": [
"威胁检测中的误报最小化",
"在资源受限硬件上的实时推理",
"上下文感知模型架构",
"操作员-AI协同与人因工程",
"多传感器融合",
"保护隐私的设备端AI"
],
"industry_preparation": {
"GitHub": "优化作品集以供国防承包商审查",
"Blog": "发布技术博客文章展示专业能力",
"Open-source": "参与与国防CV相关的开源贡献",
"Security_clearance": "安全许可申请准备事项",
"Networking": "国防科技领域人脉拓展策略"
},
"special_considerations": [
"由于训练和泰拳占用时间,可用于学习的时间有限",
"优先注重实践实现而非理论",
"聚焦战场应用场景技能",
"强调边缘部署能力",
"包含AI在战争中使用的伦理考量",
"在项目中利用美国海军陆战队(USMC)背景"
]
},
"output_format_preferences": {
"weekly_time_commitments": "明确每项活动的每周投入时间",
"prerequisites": "为每个资源标注先决条件",
"priority_levels": "标注优先级:关键/重要/有益",
"checkpoints": "每月评估进展",
"connections": "各学习路径之间的关联性",
"expected_outcomes": "每个里程碑的预期成果"
}
}分类与标签
编程开发、编程开发、学习教育、语言处理