BIKED: A Dataset for Computational Bicycle Design with Machine Learning Benchmarks

In this paper, we present “BIKED,” a dataset comprised of 4500 individually designed bicycle models sourced from hundreds of designers. We expect BIKED to enable a variety of data-driven design applications for bicycles and support the development of data-driven design methods. The dataset is comprised of a variety of design information including assembly images, component images, numerical design parameters, and class labels. In this paper, we first discuss the processing of the dataset, then highlight some prominent research questions that BIKED can help address. Of these questions, we further explore the following in detail: 1) How can we explore, understand, and visualize the current design space of bicycles and utilize this information? We apply unsupervised embedding methods to study the design space and identify key takeaways from this analysis. 2) When designing bikes using algorithms, under what conditions can machines understand the design of a given bike? We train a multitude of classifiers to understand designs, then examine the behavior of these classifiers through confusion matrices and permutation-based interpretability analysis. 3) Can machines learn to synthesize new bicycle designs by studying existing ones? We test Variational Autoencoders on random generation, interpolation, and extrapolation tasks after training on BIKED data. The dataset and code are available at http://decode.mit.edu/projects/biked/

BIKED: A Dataset and Machine Learning Benchmarks for Data-Driven Bicycle Design

In this paper, we present “BIKED,” a dataset comprised of 4500 individually designed bicycle models sourced from hundreds of designers. We expect BIKED to enable a variety of data-driven design applications for bicycles and support the development of data-driven design methods. The dataset is comprised of a variety of design information including assembly images, component images, numerical design parameters, and class labels. In this paper, we first discuss the processing of the dataset, then highlight some prominent research questions that BIKED can help address. Of these questions, we further explore the following in detail: 1) Are there prominent gaps in the current bicycle market and design space? We explore the design space using unsupervised dimensionality reduction methods. 2) How does one identify the class of a bicycle and what factors play a key role in defining it? We address the bicycle classification task by training a multitude of classifiers using different forms of design data and identifying parameters of particular significance through permutation-based interpretability analysis. 3) How does one synthesize new bicycles using different representation methods? We consider numerous machine learning methods to generate new bicycle models as well as interpolate between and extrapolate from existing models using Variational Autoencoders. The dataset is available at http://decode.mit.edu/projects/biked/ along with referenced code.

Design of an Eccentric Recumbent Ergometer to Elicit Delayed Onset Muscle Soreness

Introduction: Our objective was to design an eccentric bicycle design to elicit delayed onset muscle soreness (DOMS). Methods: To assess the bicycle designs’ ability to elicit DOMS, fourteen, recreationally active, males performed five-minutes of eccentric bicycling at 50% of their individualized power determined from a modified six-second Wingate test. Outcome measures to assess DOMS included the Likert pain scale, creatine kinase, lactate blood concentration, and pressure algometry detection evaluated at four time points (baseline (before the eccentric bicycling), immediate post, 24 hours post, and 48 hours post). Results: The Likert pain scale was different (F = 75.88, p < 0.001) at baseline (0.14 ± 0.36) and immediate post (0.21 ± 0.43), compared to 24 hours post (3.07 ± 0.83), and 48 hours post (2.93 ± 1.07). No changes were reported for creatine kinase (F = 0.7167, p = 0.475), lactate blood concentration (F = 2.313, p = 0.107), or pressure algometry detection. Conclusions: To understand mechanisms of DOMS, there is a need for a consistent, reliable method for producing DOMS. Our eccentric bicycle design and protocol offers an alternative approach to previous eccentric ergometer designs - demonstrating the potential to elicit DOMS in one, five-minute session.

Influence of Electrically Powered Pedal Assistance on User-Induced Cycling Loads and Muscle Activity during Cycling

Bicycles with electrically powered pedal assistance (PA) show great potential as ecological alternatives for engine-based vehicles. There is plenty of research available about the influence of various bicycle parameters on cycling technique. Though, to the best of the authors’ knowledge, there is none about the influence of PA. In this study, a recreational bicycle is equipped with PA and unique instrumentation to measure the user-induced loads on seat, steer and pedals. Joint loading is derived in the sagittal plane from inverse dynamics and muscle activity of the lower limbs is recorded with an electromyography system integrated in cycling pants. An experiment is set up, in which volunteers cycle on an athletics track, with a varying level of PA and a varying seat height. An ANOVA is conducted to determine significant differences due to the level of PA and seat height and to analyze the interaction effect. No interaction effect was found and only differences due to the level of PA were significant. Knowledge about the influence of PA provides insights into (i) electric bicycle design; (ii) the usage of electric bicycle for physically challenged people; (iii) the usage of electric bicycles as a rehabilitation tool.