Hi, I'm Miguel

MIT Portugal PhD Student @ U-Shift/CERIS & SIPG/ISR

What I am about.

Fourth-year MIT Portugal Transportations Systems PhD candidate at Instituto Superior Técnico. Miguel's doctoral research focuses on understanding both objective and subjective urban cycling safety and the link between the two. He is currently using Computer Vision and Machine Learning methods to create a platform for understanding both cycling safeties. He completed his MSc in Electrical and Computer Engineering at Instituto Superior Técnico and worked as a research fellow at the Institute for Systems and Robotics (ISR-IST) before starting his doctoral path.
During his free time, Miguel is 1 part Lego builder, 1 part amateur chef, and 1 part tech enthusiast.

CV (as of 31/05/2022)


Research Interests

  • Cycling
  • Safety Research
  • Urban Mobility
  • Intelligent Transportation Systems
  • Computer Vision
  • Machine Learning

Selected Projects


Cycling geo-located accidents, their details and severities

CYCLANDS is a curated collection of 30 datasets on cycling crashes to lower the barrier in objective cycling research comprising nearly 1.6M cycling accidents. This data is vital for road safety analysis, enabling researchers to develop models to understand how different factors impact the frequency and severity of accidents. Observations include the severity and location of the accident, aiming to foster the worldwide study of cycling safety by providing a testbed for researchers.

Lisbon's Circuity Analysis

Urban Street Networks,
Cycling Infrastructure Changes,

Circuity, the ratio of network distances to straight-line distances, is considered a critical measurement in urban network morphology and transportation efficiency as it can measure the attractiveness of routes in terms of distance traveled. Here, we compare circuity measures for drivable, cyclable, and walkable networks to analyze how they evolved and understand whether urban changes have produced meaningful circuity changes.


Perception of Safety,
Focus of Expansion,
Android App

Development of different metrics to assess the risk perception of a cyclist when riding a bike in an urban scenario. In this domain, the trajectory of the cyclist is considered, as well as other stationary or moving objects in its vicinity, its change in speed, acceleration and its geographic positioning (given by the smartphone) and the effort/stress of the rider, by analyzing their heart rate variability in an ECG.


Sidewalk Evaluation,
RGBD Processing

Development of a device capable of automatically digitalize and map a network of sidewalks in a city. This mapping will allow for the further development of support applications with the objective of guiding pedestrians in cities with various mobility needs. This project was funded by Thales TecInnov 1st Edition.

Bike Rider

Cycling Maneuvers,
Activity Recognition,
Machine Learning

Project to develop a sensorial map of a cyclable network supported on data captured by cyclists. To create this map we perform activity recognition. This project was funded by Thales TecInnov 2nd Edition.




  1. Costa, M., Marques, M., Roque, C., & Moura, F. (2022). CYCLANDS: Cycling geo-located accidents, their details and severities. Scientific Data, 9(1), 1-9. Link
  2. Costa, M., Félix, R., Marques, M., & Moura, F. (2022). Impact of COVID-19 lockdown on the behavior change of cyclists in Lisbon, using multinomial logit regression analysis. Transportation Research Interdisciplinary Perspectives, 100609. Link


  1. Costa, M.; Marques, M.; Moura, F. (2021). A Circuity Temporal Analysis of Urban Street Networks Using Open Data: A Lisbon Case Study. ISPRS Int. J. Geo-Inf., 10, 453. https://doi.org/10.3390/ijgi10070453 Link

Conference Papers


  1. Karina Christ, A., Costa, M., Marques, M., Roque, C., and Moura, F. (2022). Percebendo a segurança objetiva dos ciclistas urbanos: uma revisão sistemática da literatura. 10º Congresso Rodoferroviário Português, Lisboa, Portugal Link
  2. Costa, M., Marques, M., Moura, F. (2022). Como é que as redes rodoviárias, pedonais e cicláveis mudam ao longo do tempo? Uma análise da rectilinearidade em Lisboa, Portugal. 10º Congresso Rodoferroviário Português, Lisboa, Portugal Link


  1. Costa, M., Cambra, P., Moura, F., Marques, M. (2019, October). WalkBot: A Portable System to Scan Sidewalks. In 2019 IEEE International Smart Cities Conference (ISC2) (pp. 167-172). IEEE Link
  2. Cambra, P., Costa, M., Marques, M., Moura, F.(2019). WalkBot – Desenvolvimento de um Equipamento de Avaliação da Qualidade da Infraestrutura Pedonal com Recurso a Processamento de Imagem Tridimensional, in Congresso Rodoviário Português, Lisbon, Portugal Link


  1. Costa, M.; Quintino Ferreira, B.; Marques, M. (2017). A Context Aware and Video-Based Risk Descriptor for Cyclists, in 2017 IEEE 20th International Conference on Intelligent Transportation, Yokohama, Japan Link

MSc Thesis

Costa, M. (2017). Video-Based Risk Assessment for Cyclists, M.S. thesis, Dept. Elect. And Comp. Eng., UTL, Instituto Superior Técnico, Lisbon, Portugal PDF Presentation