Sampling of vehicle movement data from navigation and nomad devices is already possible and practiced. Also I discovered that Strava produces cyclist and runner heat maps online (biased). I wonder if someone has access to or published pedestrian movement data from common devices such as smart phones (less biased). I am looking to establish measures of pedestrian activity, anywhere along the road network.
Hello,
If you check my articles published here, two of them are related to pedestrian data collection using new technologies (Big Data):
https://www.researchgate.net/publication/293605778_Big_data_helps_pedestrian_planning_take_a_big_step_forward
https://www.researchgate.net/publication/293605901_Understanding_How_Big_Data_and_Crowd_Movements_Will_Shape_the_Cities_of_Tomorrow
As I mention, there are different ways to collect data for pedestrians, but nowadays mobile phone data is not the answer. There are some limitations in the use of mobile data even to track vehicles (I have another article about that), so imagine the limitations in the use of mobile data from operators to track pedestrians... Some telecom companies provide mobile data, but it is very important to know about the limitations, and these are the main ones:
1) mobile data work by using the antennas from the operators to track the movement of people. It could be enough to track vehicles (from one part of the city to the other one) but not for short-distance trips. Pedestrians wont walk more than few kilometres, and they will be using the same antenna... so it is impossible, nowadays, to accurately track pedestrians movements through the city by using mobile phone data from operators.
2) In city centres, pedestrian may have the same speed as buses, cars, bikes... therefore, the cluster analyses wont correctly differenciate between pedestrians or other modes.
3) Mobile data from operators may be also biased. A big operator (Vodafone, O2...) will have about 20-30% of the market share. That means that they will need to expand the data to the whole population. But the market share varies by zone, so the data will be biased also and you will have to take into account these expansion errors when using it.
Therefore, mobile phone data is not the best option to capture pedestrians. You could use it to capture longer trips (for example cars), but also with limitations, or to know the number of people arriving at the city centre, but with no details about routes, mode of transport...
Please ask me if you need more information.
Thanks!
Article Big data helps pedestrian planning take a big step forward
Conference Paper Understanding How Big Data and Crowd Movements Will Shape th...
Dear Chris,
I provide a review of data sources for pedestrians in my thesis published in 2015 (Section 2.1, p.9), including several data collection techniques (Manual counts and shadowing, Mechanical counts, Survey data, Lasers and infrared sensors, Video counting, GPS, RFID tags and readers, smart card, Data from the mobile phone). I specifically focus on Mobile phone tracking data from antennas. I also provide information about map data, since they are also very useful. Check here: http://dx.doi.org/10.5075/epfl-thesis-6806
I used WiFi traces from access points in my thesis and the raw data (i.e., WiFi data from access points, aggregate occupancy data for each point of interest and map data) are available here: http://dx.doi.org/10.5281/zenodo.15798 (please cite my paper in Transp. Res. Part C when using these data, since they are associated to this paper: http://dx.doi.org/10.1016/j.trc.2014.03.015).
We also used data from the WiFi sensor on mobile phones (data from the device, as opposed to the previous paragraph, data from the antennas) in this Master thesis: http://transp-or.epfl.ch/documents/masterTheses/BUISSON14.pdf if you're also interested in this kind of data.
I hope it helps! Don't hesitate to contact me if you have questions.
Best regards,
Antonin
Hello!
I would like to suggest you this "brand new" PhD thesis discussing also this subject:
http://infoscience.epfl.ch/record/214544/files/EPFL_TH6806.pdf
I have personally done some work to use pedestrian trajectories to calibrate cellular automaton models. Maybe it could be interesting to you for the future.
https://www.researchgate.net/publication/279962882_Calibrating_floor_field_cellular_automaton_models_for_pedestrian_dynamics_by_using_likelihood_function_optimization
Regards
Rino
Article Calibrating floor field cellular automaton models for pedest...
Wow great feedback. I feel like I learning a lot in this space. Thanks to all who answered.
Additional interesting research, including automated pedestrian trajectory data-collection - http://n.saunier.free.fr/saunier/
Looks like this is an interesting question, with lots of development occurring, fast. I have read extensively and see a way forward to answer this research question. I have 'shelved' this research due to lack of funding for the moment. There was no demand for this knowledge among my regular research funding partners (road agencies). Hopefully, we can attract either a new source of funding, or the existing funding bodies engage more deeply in active transport.
Not sure if this data source is detailed enough but https://besttime.app provides foot traffic data for individual stores across the world, based on location data collected from mobile phone apps running in the background.
- Badges
- Science topic
- Similar topics
- Biological Science
- Microbiology
- Virology
- Virus