Real-World Robotics – A Hands-On Project Class

Implementation of the Project/Course

The course revolves around the task of designing and building a robotic hand from scratch to complete a set of autonomous challenges. Students must move through the full development process—from initial design to fabrication, programming, and testing—giving them an end-to-end perspective rarely offered in traditional coursework.

Teaching mode and format
The course is taught fully in-person, structured around a blended-learning approach that combines asynchronous preparation with synchronous, hands-on sessions. Each week begins with a short tutorial video followed by a Focus Talk where instructors expand on the material and address student questions. Workshops led by PhD students and teaching staff provide direct exposure to implementation methods, after which teams apply the material to their own systems.

Active learning and feedback
The emphasis is on active, project-based learning. While short lectures provide theoretical input, the majority of learning takes place through team work, workshops, and continuous iteration on the robots. Feedback is given at multiple levels: informally through Slack, during weekly check-ins with TAs, and through structured milestones. A final demonstration at the end of semester provides summative feedback.

Engagement and communication
Engagement is extremely high: students regularly work late into the night to iterate on their designs and test systems. A dedicated Slack workspace has proven invaluable, enabling rapid communication between students and staff and reducing barriers to asking questions. Peer-to-peer exchange is also strong, with students sharing insights and solutions across teams. For staff, Slack and regular coordination meetings ensure that feedback and supervision are aligned.

Synchronous and asynchronous elements
Tutorial videos and quizzes on Moodle provide asynchronous learning opportunities, allowing students to prepare and review at their own pace. The synchronous components—Focus Talks, workshops, and hands-on sessions—emphasize interaction, collaboration, and direct application.

Student support offerings
In addition to workshops and TAs’ presence in the lab, regular hardware/software check-ups are built into the semester to prevent bottlenecks and ensure that teams can progress. This combination of proactive support and on-demand help has been highly valued by students.

Assessment strategy
Teams are evaluated against a clear rubric rather than relative performance. Assessment criteria include technical functionality, creativity, teamwork, and project documentation. This approach avoids unhealthy competition, fosters camaraderie, and ensures fairness.

Challenges and lessons learned
The biggest challenge is balancing ambition with workload: building a complex robot in one semester is demanding, and students often experience steep learning curves. To address this, we introduced stronger

scaffolding in the early weeks, clearer milestones, and more structured support. Another challenge is resource availability (e.g., limited 3D printers or motors), which we mitigate through careful scheduling and recycling of components.

Overall, the implementation fosters high levels of engagement, teamwork, and knowledge transfer. Students not only learn advanced robotics concepts but also how to collaborate effectively under real-world constraints. This balance of structured guidance and open-ended project work has been central to the course’s success.

Motivation, Project Mission, Vision Statement

The purpose of the project is to equip students with the skills to address real-world robotics challenges in both industry and academia. While students often study robotics theory (e.g., controls, dynamics, planning, sensing), they rarely apply it to a real robotic system, let alone in an end-to-end manner. RWR closes this gap by giving students the chance to design, build, and control a robot from start to finish, bridging theory with practice.

The course is guided by two core values: (1) learning by doing and (2) teamwork. The gap between theory and practice is wide, and its nuances can only be understood through experience. Robotics is also inherently interdisciplinary, requiring collaboration across many domains. The course is structured so that each student’s contribution is essential to the team’s success, reinforcing accountability and collective problem-solving.

The goals of the project are to:

– Develop practical and theoretical skills in parallel while addressing implementation bottlenecks.

– Solve real robotics problems to deepen understanding and build critical perspectives.

– Cultivate competence in project management and organizing a development process efficiently as a team.

– Expose students to state-of-the-art methods and tools used in modern robotics.

Innovative Elements

A key innovation of the project is that students build a complete robotic system from start to finish, rather than only learning isolated parts. Working in balanced teams around a concrete challenge, they move through the entire process—from initial sketches to autonomous behaviors on real robots. Furthermore, the course combines modern engineering tools like CAD design and 3D printing with cutting-edge methods such as reinforcement and imitation learning for autonomy. Students also explore new interfaces like motion capture gloves and robotic arms for direct control. This blend of hands-on building, advanced methods, and teamwork differs from traditional courses and gives students a unique end-to-end experience.

Effects on Student Learning

Students’ engagement was evident throughout the semester: they could often be found working late into the night, iterating on designs, collecting data, and seeking feedback during check-ins or on Slack. As one student noted, “The output we reached in 2.5 months is unbelievable, we had the possibility to develop from scratch an entire and complex project like this.” The impact is also visible in outcomes: since 2023, student-led projects from the course have resulted in 3 published papers at IROS, one of the leading international robotics conferences, with one design open-sourced and now used by research groups worldwide. Alumni have gone on to theses at Stanford and Harvard and internships at companies such as Tesla. Demand has steadily grown, with applications rising from 24 in 2023 to over 60 in 2025, reflecting the course’s effectiveness and student recognition of its value.

Which Elements of Your Project Would You Recommend to Others?

One element we strongly recommend is evaluating teams against a clear rubric rather than in direct competition. This fosters collaboration and mutual support across the course, building camaraderie rather than rivalry. Another transferable element is structuring projects so that success depends on contributions from all team members, reinforcing accountability and teamwork. We achieve this during the team formation phase, by grouping students based on their goals and balancing skills across teams. While this approach requires more work from the teaching staff, we have found it creates a more motivating and supportive learning environment.