Paper here.

My PhD robot.

Summary:

It presents the result of a project which wishes to demonstrate the feasibility of an affordable humanoid robot named R1. It can navigate, and interact with the environment (grasping and carrying objects, operating switches, opening doors etc).

Healthcare: R1 could be used to assist nurses and doctors with tasks such as patient care, medication delivery, and transportation.

Elderly care: R1 could provide companionship and assistance to elderly people in their homes, helping them with tasks such as meal preparation, cleaning, and bathing.

Education: R1 could be used as a teaching assistant in schools, helping teachers to engage students and provide them with personalized instruction.

The expected cost of R1 is around that of a family car, and possibly even significantly lower when produced in large quantities.

Paper here.

An easy way to evaluate your robotic hands.

Summary:

We propose a new evaluation index for robotic hands called the FFP Index, based on three key areas: form, features, and performance.

Form: How anthropomorphic is the hand in terms of its basic mobility and appearance?

Features: What features does the hand have that allow for effective control?

Performance: How well does the hand perform on a series of tasks?

We also include price as an additional feature to provide context when comparing multiple hands.

Paper here.

A new approach to robot design is cascaded optimization, which leads to more novel and better-performing robots.

Summary:

Standardized robot modules often derive their characteristics from conventional industrial robots, making them expensive, bulky, and potentially limiting their wider adoption. To address this issue, a new top-down multidisciplinary computational design strategy that:

Identifies non-intuitive designs: Cascaded optimization can identify robot architectures that would not be considered since it optimizes all of the design variables together, rather than separately.

Recovers lost performance: It can recover some of the performance that is lost when deviating from the computed optima to ensure physical feasibility.

Leads to better-performing robots: Cascaded optimization can lead to more synergistic designs.

It is a promising new approach to robot design that has the potential to revolutionize the way that robots are built.

Paper here

Trash robots that aren’t so trash!

Summary:

We developed an autonomous trash-picking robot that elevates the importance of human-robot interaction (HRI) within its design process utilizing a modified double diamond framework.

HRI-Centric Design: The robot incorporates HRI-focused hardware features, enhancing the user experience by ensuring ease of operation.

User-Centric Methodology: The design process involved ideation workshops, storyboarding, and mood boards to align the robot's functionalities with user requirements and preferences.

Practical Implementation: The robot was tested in various scenarios, confirming its capability to perform its primary function of detecting and collecting small trash efficiently.
The robot combines technical functionality with a strong emphasis on user-friendly design and interaction. It stands as a testament to the potential of integrating HRI principles into the core of robotic design to meet practical needs in public and private spaces.

Paper here

Our first attempt at optimizing a humanoid robot arm topology.

Summary:

We suggest the potential for improved stiffness budget allocation and the inclusion of dynamic load cases to refine the design methodology.

Innovative Approach: This paper introduces a modular approach to the topology optimization of a humanoid robot arm, by optimizing each component separately, the study addresses the challenge of excessive computational demands typically associated with multi-component robotic systems.

Material Analysis: The research explores the use of different materials—ABS plastic, aluminum, and titanium—for the arm's segments.

Future Directions: The paper concludes with insights into future enhancements, such as integrating the optimization results into the physical design process to ensure mass efficiency in the manufactured structure.

This work paves the way for more efficient, lightweight robotic designs that can perform a variety of tasks with reduced energy consumption and increased agility which we explored in following papers.

Paper here

Our work on integrating TMDs into optimized lattice structures for vibration reduction.

Summary:
The paper introduces a groundbreaking method for enhancing the vibrational characteristics of structures by embedding multiple tuned mass dampers (m-TMDs) within the cavities of topology-optimized unit cells.

Innovative Integration: We effectively reduce vibrations without the need for additional space.

Optimization Technique: A unique physics-informed penalty factor tailored to the chosen unit cell is utilized for topology optimization.
Dynamic Performance Improvement: The application of this technique to a 2-segment robot arm results in a unit cell robotic arm (UC-Arm) that is not only 3.6% lighter than the reference model but also exhibits a 60% reduction in dynamic displacement.

This research showcases the potential for advanced design strategies in robotics, where dynamic and static requirements are addressed separately, leading to components that are both weight-efficient and exhibit superior dynamic load handling.

Paper here

Our approach to a bioinspired design workflow.

Summary:
The chapter outlines a comprehensive workflow that seamlessly integrates bioinspired concepts with classical design methodologies.

Unified Design Workflow: This workflow is adaptable for both problem-driven and solution-driven strategies, ensuring versatility in tackling design challenges.

Bioinspired Tools and Techniques: It introduces structural and functional biocards as pivotal tools for problem formulation and abstraction, which are crucial in translating biological phenomena into viable engineering solutions.

Practical Application and Validation: The workflow's practicality is demonstrated through its application in an educational setting, where it guided student teams in successfully developing prototypes.

This chapter contributes a structured approach to the field of bioinspired design, offering a replicable and validated method for developing innovative solutions inspired by nature.

Paper here

Our framework for enhancing start-up product development.

Summary:

The paper builds on previous research by refining the role-based prototyping process, which is crucial for start-ups to efficiently transform ideas into marketable products.
Role-Based Prototyping Enhancement: This enhancement includes a prototyping matrix to guide resource allocation and focus.

PETRA Framework Application: The PETRA (Plan, Execute, Test, Reflect, Assimilate) framework is introduced, providing a structured cycle for prototyping, along with a modified Kanban board, Protoban, for managing the development process.

Focused and Efficient Development: It aids teams in understanding what actions to take and when, optimizing team dynamics and progress tracking, ultimately leading to a more effective path to a minimum viable product.

The work showcases how structured frameworks like PETRA can significantly support start-ups in navigating the early stages of product development, leading to better resource management and a clearer development trajectory.

Paper here

Our computational approach to design low-cost, lightweight robots.

Summary:

This paper outlines a computational design strategy for creating modular robots that are both affordable and lightweight, which is crucial for customization in industrial and personal applications.

Design Exploration: The study begins with a search strategy to explore various robot module designs and is crucial for identifying non-standard robot structures that can meet specific task requirements.

Weight Reduction: Following the architectural design, a dynamics-informed structural optimization to minimize the robot's weight leading to structural mass being 16% lighter compared to those designed with conventional materials, like aluminum tubes.

Feasibility : The designed modules ensure the robots' physical feasibility, with potential directions including the integration of 3D-printed components and aluminum tubes for an optimal balance of cost and functionality.

The proposed design strategy can be applied to a broader range of modular robots, potentially influencing industry standards for robot design.