Overview
Advances in artificial intelligence and robotics have increasingly relied on better algorithms or larger datasets while often overlooking the critical role of hardware. In this workshop, we aim to addresses the wide gap between mechanical engineers, who design complex and innovative hardware, and computer scientists, who develop intelligent software for simplified or standard platforms. Bridging this divide is essential for creating integrated, intelligent systems that fully exploit the potential of both fields. By bringing together researchers from robotics, machine learning, and mechanical engineering, in this workshop we aims to foster interdisciplinary collaboration, present cutting-edge research, and propose new methodologies for unifying hardware and intelligence in robotics.🏆 Awards
🎓 Best Workshop Paper Award 🎓
AnySkin: Plug-and-play Skin Sensing for Robotic Touch
🤖 Best Workshop Demo Award 🤖
Touch in the Wild: Learning Fine-Grained Manipulation with a Portable Visuo-Tactile Gripper
Areas of Interest
We invite contributions and discussions on topics related to hardware-aware intelligence, including:
- Hardware for Sample Efficient Learning: Exploring how novel hardware designs facilitate sample-efficient learning-based control or support data collection.
- Hardware for Low Cost/Scalable Data Collection: Developing intelligent robotic systems using low-cost components, open-source hardware, and rapid prototyping tools to democratize robotics research and enable scalable data collection and experimentation.
- Sensor Optimization: Designing, placing, and integrating sensor suites to maximize perceptual fidelity and improve learning outcomes.
- Generative and Evolutionary Hardware Design: Leveraging generative models, evolutionary algorithms, and data-driven techniques to automatically discover and optimize robotic structures and control strategies.
- Co-Design of Control and Hardware: Joint optimization of physical design and control policies for high-performance robotic systems.
- Manufacture-in-the-Loop Design and Fabrication: Integrating manufacturing constraints and capabilities directly into the design loop to enable iterative, fabrication-aware robot design.
- Learning for Unconventional Hardware Platforms: Developing modeling and learning methods tailored to unconventional hardware, such as soft robots, underactuated mechanisms, and reconfigurable structures.
- Foundation Models for Hardware Design: Exploring the role of large-scale pretrained foundation models in automating and accelerating robot design, simulation, and optimization workflows.
Invited Speakers
Tentative Schedule
Call for Papers and Demos
We invite both paper submissions and hardware demo submissions. Submissions should follow the guidelines below:
-
Paper Submissions (Poster Presentation + Selected Lightning Talks):
We invite 4–8 page papers in the RSS 2025 paper format. Submissions will be single-blind. All accepted papers will be presented as posters during the workshop. A subset of papers will be selected for 3-minute spotlight talks. References and supplementary materials are not counted towards the page limit.
Submission Link: OpenReview submission
-
Hardware Demo Submissions:
We encourage the submission of hardware demonstrations, which will be presented interactively at the workshop to foster hands-on engagement with attendees. Submissions will be single-blind and must include:
- A short paragraph describing the system
- A description of how attendees will interact with the system
- Multimedia (e.g., video or images) showcasing the system in action
- A list of resources required (e.g., space, internet, electrical outlets, etc.)
No camera-ready version is required. Commercial sales and marketing activities from companies are allowed.
Submission Link: Hardware Demo Submission Form
Organizing Committee
| Huy Ha | huyha@stanford.edu | Stanford University |
| Zhanyi Sun | zhanyis@stanford.edu | Stanford University |
| Xiaomeng Xu | xuxm@stanford.edu | Stanford University |
| Yuxiang Ma | yxma20@mit.edu | MIT |
| Sha Yi | shayi@ucsd.edu | UCSD |
| Mike Tolley | tolley@ucsd.edu | UCSD |
| Xiaolong Wang | xiw012@ucsd.edu | UCSD |