主办：陕西省汽车工程学会
ISSN 1671-7988  CN 61-1394/TH
创刊：1976年
汽车实用技术

汽车实用技术 ›› 2025, Vol. 50 ›› Issue (20): 134-138.DOI: 10.16638/j.cnki.1671-7988.2025.020.025

• 汽车教育 • 上一篇    

基于百度 Apollo 平台的高校自动驾驶教学创新

郑竹安，朱乃宣，谢双健   

  1. 盐城工学院 汽车工程学院
  • 发布日期:2025-10-17
  • 通讯作者: 郑竹安
  • 作者简介:郑竹安（1984－），男，博士研究生，副教授，研究方向新能源汽车先进底盘技术及无人驾驶技术
  • 基金资助:
    教育部产学合作协同育人项目（231100805071631）；教育部供需对接就业育人项目（2024010419776）；校 产教融合型一流课程（CJRH202224）；盐城工学院“人工智能+高等教育”典型应用场景案例

Innovative Autonomous Driving Teaching in Universities Based on the Baidu Apollo Platform

ZHENG Zhu'an, ZHU Naixuan, XIE Shuangjian   

  1. School of Automotive Engineering, Yancheng Institute of Technology
  • Published:2025-10-17
  • Contact: ZHENG Zhu'an

PDF

摘要： 为落实《教育强国建设纲要》中“以教育数字化开辟发展新赛道”的战略要求，针对 当前高校自动驾驶教育面临的技术滞后、实践缺失、产教脱节等核心痛点，文章提出基于百 度阿波罗（Apollo）平台构建“开放平台+课程体系+产业场景”的智能驾驶教育生态（ApolloX）。该生态通过“仿真教学+实车验证+产业对接”三维模式，整合 Apollo 自动驾驶仿真工具 链与机器人操作系统（ROS）智能车平台，开发跨平台算法迁移框架与数字孪生教学环境， 实现理论学习、仿真训练、实车验证到产业落地的全链路培养。实践表明，该生态已在全国 50 余所高校落地应用，支撑 20 余个标准化课程模块建设，孵化 100 余项学生创新项目，实 验教学成本降低 83%，实验资源利用率提升 340%，师资培训周期缩短 60%，显著提升了学生 工程实践能力与行业适配度，为智能交通领域新型人才培养及教育数字化转型提供了可复制 的范式。

关键词: 自动驾驶；Apollo；ROS；实践

Abstract: To implement the strategic requirement of "leveraging educational digitization to pioneer new development pathways" outlined in the Outline for Building a Leading Educational Nation, this paper addresses three critical challenges in current autonomous driving education in higher education institutions–technological obsolescence, lack of practical training, and disconnection between industry and academia–by proposing the Apollo-X intelligent driving education ecosystem. This ecosystem, based on Baidu's Apollo platform, integrates an "open platform+curriculum system+industrial scenarios" framework. Adopting a three-dimensional model of "simulation-based teaching+real vehicle validation+industrial collaboration," the ecosystem combines Apollo's autonomous driving simulation toolkit with the robot operating system (ROS)-based smart vehicle platform. It features a cross-platform algorithm migration framework and a digital twin teaching environment, enabling end-to-end talent development from theoretical learning and simulation exercises to real-vehicle testing and industrial implementation. Practical results demonstrate that the ecosystem has been deployed in over 50 universities nationwide, supporting the development of 20+ standardized course modules and incubating 100+ student innovation projects. It has achieved an 83% reduction in experimental teaching costs, a 340% improvement in resource utilization efficiency, and a 60% shortening of faculty training cycles. These outcomes have significantly enhanced students' engineering competencies and industry readiness, establishing a replicable paradigm for cultivating next-generation talent in intelligent transportation and advancing the digital transformation of education.

Key words: autonomous driving; Apollo; ROS; practice