课程简介
《ME336协作机器人学习》是一门面向南方科技大学机器人工程专业大三、大四年级开授的本科生专业选修课,参课学生也包括来自机械工程、力学航空与航天工程、电子工程、计算机科学与工程等相关专业本科生。由南方科技大学机械与能源工程系助理教授宋超阳博士担任责任教师,是国内首门以机器人学习为主题的机器人工程专业课程,通过多学科交叉的方式,融合了以协作机器人为代表的新兴机器人硬件系统、以机器学习为引领的人工智能技术、以及以「项目实践-客座报告-教具研制-线上同步-双语教学」相结合的新工科教学方法创新。
The “ME336 Collaborative Robotics Learning” is an undergraduate elective course offered to third and fourth-year students majoring in Robotics Engineering at Southern University of Science and Technology (SUSTech), with participants also including undergraduates from related majors such as Mechanical Engineering, Mechanics, Aeronautics and Astronautics Engineering, Electronic Engineering, Computer Science and Engineering. Dr. Song Chaoyang, an Assistant Professor in the Department of Mechanical and Energy Engineering at SUSTech, serves as the lead instructor. This course is the first in China to focus on robotics learning in the field of robotics engineering. It integrates emerging robotic hardware systems represented by collaborative robots, artificial intelligence technologies led by machine learning, and innovative engineering education methods combining “project practice, guest lectures, teaching aid development, online synchronization, and bilingual teaching” through interdisciplinary integration.
教学特色
该课程目前作为「南方科技大学机器智能设计与学习虚拟教研室」的核心课程之一,经过多年迭代发展,已经逐步实现教学内容、教学方法、教学工具、教学团队等多方面的迭代与优化,致力于借助设计科学方法构建一套与参课同学交流反馈并深度合作为核心的南科大特色「跨课程、连贯式、大团队」新工科教改成果。
Currently, as a core course of the “SUSTech Virtual Research and Teaching Group for Intelligent Design and Learning of Machines,” the course has undergone iterative development and optimization in teaching content, methods, tools, and team over the years, aiming to build a SUSTech-specific “cross-course, coherent, large-team” engineering education reform achievement centered on deep cooperation and communication with participating students through design science methods.
教学成果
自2019年春季首次授课至今,参课学生人数逐年提升并实现了6倍增长。基于该课程研制的DeepClaw柔性力触觉便携式教具系统(具体可参见https://deepclaw.ancorasir.com/),于2023年12月经过全球评选,获得由联合国教科文组织高等教育研究中心首届颁发的《高等教育数字化先锋案例奖》。
Since its first offering in the spring of 2019, the number of students enrolled in the course has increased sixfold. The DeepClaw flexible force tactile portable teaching aid system developed based on this course (see https://deepclaw.ancorasir.com/) was awarded the “Higher Education Digitalization Pioneer Case Award” by the UNESCO Higher Education Research Center in December 2023 after global selection.
客座报告
本课程在教学过程中,积极邀请国内外来自产业界以及学术界人工智能与机器人技术的青年专家与学者,走进南科大课堂,每年通过ME336课程开展《机器人与人工智能客座报告论坛》,让参课学生可以通过客座专家的学术与技术报告内容、结合其专业学习与职业成长经历,进一步了解课堂以外的机器人学习领域前沿应用案例与技术发展方向,为同学们提供一个围绕机器人学习为主题的全方位视角。
In the teaching process, the course actively invites young experts and scholars from the industry and academia in artificial intelligence and robotics technology from both domestic and international communities to enter SUSTech classrooms. Through the ME336 course, the “Robotics and Artificial Intelligence Guest Lecture Forum” is held annually, allowing participating students to further understand the cutting-edge applications and technological developments in the field of robotics learning outside the classroom through the academic and technical reports of guest experts, combined with their professional and career growth experiences, providing a comprehensive perspective on robotics learning.
教具研发
机器人学习作为一门将机器学习与机器人技术相结合前沿学科,是目前国际学术研究的前沿热点,也是企业招聘时需求强烈的人才培养重要方向。需要同学们交叉融合人工智能与机器人学两个学科方向的基础理论与前沿技术,同时针对来自现实生活与企业研发过程中出现的多方面应用挑战,进行多学科交叉融合的知识学习与技能掌握。此前在诸如美国斯坦福大学、加州大学伯克利分校、德州大学奥斯汀分校、德国马普所人工智能研究院、汉堡大学等院校先行开授过以机器人学习为主题的实验性课程,但在国内尚属空白。
Robotics learning, as a frontier discipline combining machine learning and robotics technology, is a current hot topic in international academic research and an important direction for talent cultivation in high demand by enterprises. Students are required to integrate the basic theories and cutting-edge technologies of artificial intelligence and robotics, and to master interdisciplinary knowledge and skills to address various application challenges in real life and enterprise R&D processes. Experimental courses on robotics learning have been offered in institutions such as Stanford University, University of California, Berkeley, University of Texas at Austin, Max Planck Institute for Intelligent Systems, and University of Hamburg, but such courses are still a new field in China.
教学模式
这些课程一方面为《ME336协作机器人学习》课程的顺利开授提供了宝贵的教学经验,同时也对开授这门新课提出了诸多挑战,包括缺少规范的机器人学习专业教材、需要结合中英文双语开展理论加实践的教学过程、需要计算硬件与机器人硬件相结合的实验教学场地与设备、需要一群可以在机器学习与机器人技术方面互为补充的教学支撑团队等。通过近年来的ME336课程迭代以及与每届参课同学的交流反馈,特别提出并实践了「积极吸引上一届参课同学配合课程指导老师、实质性的参与到下一届课程实践教学环节的课程优化设计」的南科大特色教改方案,在多个南科大教改项目的支持下,逐步形成了目前的《ME336协作机器人学习》课程教学模式(具体可参见https://me336.ancorasir.com/)。
These courses have provided valuable teaching experience for the successful offering of the “ME336 Collaborative Robotics Learning” course, while also presenting many challenges, including the lack of standardized textbooks for robotics learning, the need for bilingual teaching combining theory and practice, the need for experimental teaching facilities and equipment combining computing hardware and robotic hardware, and the need for a teaching support team that can complement each other in machine learning and robotics technology. Through the iteration of the ME336 course in recent years and the communication and feedback from each cohort of participating students, the SUSTech-specific education reform proposal of “actively attracting students from the previous cohort to substantially participate in the next cohort’s practical teaching under the guidance of course instructors” has been proposed and practiced. With the support of various SUSTech education reform projects, the current teaching model of the “ME336 Collaborative Robotics Learning” course has been gradually formed (see https://me336.ancorasir.com/).
课程项目
该课程包含如下四个方面的培养目标,包括1)学习使用基于视觉的机器学习和人工智能技术、在现实世界中构建机器人系统的基础知识;2)了解构建基于学习的机器人操控系统所面临的技术挑战;3)熟悉机器人学习中各种模态驱动和数据驱动的原理和算法;4)能够评估、沟通和应用基于人工智能的技术解决机器人学中的问题。
The course has the following four training objectives: 1) Learning to use vision-based machine learning and artificial intelligence technologies, and acquiring fundamental knowledge for building robotic systems in the real world; 2) Understanding the technical challenges faced in constructing learning-based robotic manipulation systems; 3) Familiarizing with various modality-driven and data-driven principles and algorithms in robotics learning; 4) Being able to evaluate, communicate, and apply AI-based technologies to solve problems in robotics.
参加这门课程的同学需要具备多方面前期基础,包括1)基础数据结构和算法知识,以及计算机编程的实践技能;2)精通Python是必需的,对C/C++有较高水平的熟悉度是一个加分项;3)熟悉微积分、统计学和线性代数、需要强大的数学技能;4)优先考虑在机器人学、机器学习和人工智能方面的课程学习和/或相当经验。
Students participating in this course need to have a multifaceted foundation, including 1) basic knowledge of data structures and algorithms, as well as practical skills in computer programming; 2) proficiency in Python is required, and a higher level of familiarity with C/C++ is a plus; 3) familiarity with calculus, statistics, and linear algebra, requiring strong mathematical skills; 4) priority consideration for coursework and/or equivalent experience in robotics, machine learning, and artificial intelligence.
在课程学习过程中,除核心理论与技术,同学们还需要综合理解并实践多方面的技能,包括:1)根据功能和环境要求,在团队设置中利用概念生成方法,达成机器人设计概念的共识;2)从功能、硬件、算法和物理环境的角度设计和开发功能性机器人程序;3)应用包括机械工程、电气工程、应用数学和计算机科学在内的学科基础,理解机器人的智能化应用实现;4)在多媒体展示和报告写作中沟通并形成科学有效的工程决策,并通过设计、编程和测试结果进行试验性验证。
During the course, in addition to core theories and technologies, students also need to comprehensively understand and practice various skills, including: 1) using concept generation methods in team settings to reach consensus on robotic design concepts based on functional and environmental requirements; 2) designing and developing functional robotic programs from the perspectives of functionality, hardware, algorithms, and physical environment; 3) applying disciplinary foundations including mechanical engineering, electrical engineering, applied mathematics, and computer science to understand the intelligent application implementation of robots; 4) communicating and forming scientifically effective engineering decisions in multimedia presentations and report writing, and experimentally validating them through design, programming, and testing results.
学生获奖
参课同学依托课程项目学习内容,先后获得获得南方科技大学工学院综合设计项目一等奖、二等奖、三等奖各1项、2023年中国大学生机械工程创新创意大赛机械产品数字化设计赛一等奖、第六届中国国际“互联网+”大学生创新创业大赛校级三等奖、第十二届“挑战杯”广东大学生创业大赛大学生创业计划竞赛省部级二等奖、第十四届“挑战杯”广东大学生创业大赛大学生创业计划竞赛校级二等奖等多个奖项。
Based on the course projects, students have won various awards including the First, Second, and Third Prizes of the Comprehensive Design Project of the School of Engineering at Southern University of Science and Technology, the First Prize of the Digital Design Competition of Mechanical Products in the 2023 China College Student Mechanical Engineering Innovation and Creativity Competition, the Third Prize of the School-level Competition in the Sixth China International “Internet+” College Student Innovation and Entrepreneurship Competition, the Second Prize of the Provincial-level Competition in the Twelfth “Challenge Cup” Guangdong College Student Entrepreneurship Competition, and the Second Prize of the School-level Competition in the Fourteenth “Challenge Cup” Guangdong College Student Entrepreneurship Competition.
学生论文
依托课程项目学习内容,参课同学已经在包括The International Journal of Robotics Research、Advanced Intelligent Systems、Materials & Design、IEEE Robotics and Automation Letters等高水平学术期刊以及Conference on Robot Learning (CoRL)、AIM、CASE、RoboSoft、ICARM等机器人方向国际学术会议上发表多篇学术研究论文,获得AIS期刊封面论文2次、会议最佳论文奖2次。
Leveraging the course project learning content, participating students have published several academic research papers in high-level academic journals such as The International Journal of Robotics Research, Advanced Intelligent Systems, Materials & Design, IEEE Robotics and Automation Letters, and international academic conferences in robotics such as Conference on Robot Learning (CoRL), AIM, CASE, RoboSoft, and ICARM. They have received two cover papers in AIS journal and two Best Paper Awards at conferences.
教研融合
「南方科技大学机器智能设计与学习虚拟教研室」由南方科技大学机械与能源工程系助理教授宋超阳博士牵头组建,核心成员还包括机器人研究院院长戴建生院士、王峥教授、助理教授潘阳博士、助理教授贾政中博士、研究副教授刘思聪博士等,依托所教授的机械设计、高等机构学及其应用、软体机器人、先进机器人驱动技术、机器人建模与控制等多门关联领域核心课程的教学经验与教学内容,联合开展教学与科研相结合的新工科教学改革与合作。虚拟教研室成员此前也获批了包括南方科技大学-麻省理工学院机械工程研究与教学联合中心项目《Interactive Human-Robot System for Robotics Education and Research》、南方科技大学工学院教学创新项目《协作机器人用于新工科互动教学》、南方科技大学教育教学研究和改革项目《跨课程连贯式大团队实验项目新工科教改》等多个人工智能与机器人新工科教学改革项目。虚拟教研室的代表性成果包括2023年12月获得由联合国教科文组织高等教育研究中心首届颁发的《高等教育数字化先锋案例奖》,同时也负责南方科技大学机器人社团的竞赛指导,组织感兴趣的同学参加包括RoboCon、全国大学生机械设计大赛等多项高水平赛事。
The “SUSTech Virtual Teaching Group for Machine Intelligence Design and Learning” is led by Dr. Song Chaoyang, Assistant Professor in the Department of Mechanical and Energy Engineering at Southern University of Science and Technology, with core members including Academician Jian Sheng Dai, Dean of the Robotics Institute, Professor Zheng Wang, Assistant Professor Yang Pan, Assistant Professor Zheng Zhong Jia, and Associate Research Professor Si Cong Liu. Based on the teaching experience and content of core courses in related fields such as mechanical design, advanced mechanism theory and its applications, soft robotics, advanced robotic actuation technologies, and robotic modeling and control, the group collaboratively conducts new engineering education reforms and cooperation, combining teaching and research. The virtual research and teaching group has also been approved for several AI and robotics new engineering education reform projects, including the “Interactive Human-Robot System for Robotics Education and Research” project in the SUSTech-MIT Joint Center for Mechanical Engineering Research and Teaching, the “Collaborative Robots for Interactive New Engineering Education” project in the SUSTech School of Engineering Teaching Innovation Program, and the “Cross-Course Coherent Large-Team Experimental Project New Engineering Education Reform” project in the SUSTech Education and Teaching Research and Reform Program. The representative achievements of the virtual research and teaching group include receiving the “Higher Education Digitalization Pioneer Case Award” issued by the UNESCO Higher Education Research Center in December 2023, and also being responsible for the competition guidance of the SUSTech Robotics Club, organizing interested students to participate in high-level competitions including RoboCon and the National College Student Mechanical Design Competition.