According to their case studies, they mainly serve agencies that manage public lands.
The TRAFx Trail Counter uses infrared to measure people on trails, paths, and sidewalks. It uses 3 AA batteries and can last for 10 years.
The TRAFx Vehicle Counter uses a magnetometer to measure passing vehicles. It has a Mountain Bike Counter mode that probably dials up the sensitivity. It uses 3 C that last 8-9 months or 20 months with a Lithium D battery. There’s no mixed use roadway mode, so cars will cause miscounts on a counter in mountain bike mode.
The entire system, including 3 of their counters costs $2215. With a system package, additional counters cost $460 to $540.
One of the first search hits for bicycle counting systems is a French company called Eco-Counter. The Multi-nature 2 counts pedestrians, bikes, equestrians, and ATVs. However, it requires more than just a passive IR sensor (called PYRO) and calls for installing their inductive loop system (called ZELT) underneath the trail surface. These devices last for 2 years before needing to be recharged.
I can’t find out how much these systems would cost and, unlike TRAFx, appear to be a permanent installation.
The TRAX Apollyon Plus II is a tube counter module, introduced in 2018, invoiced to the Texas DOT at $1,195 each in April 2019. It’s semi-permanent, requiring a chain and lock.
The TDC-Ultra for Bicycles is a manual counter panel with buttons to distinguish attributes like gender and the presence of a helmet.
Q-Free is a Norwegian company that sells the HI-TRAC CMU bicycle and pedestrian counting system, which combines a piezoelectric sensor for bikes and a pyroelectric infrared sensor for pedestrians. The raw data is passed to noise filtering and estimation heuristics that improve the accuracy of the sensor. There’s an active signage model that displays counts to passersby.
It seems to be focused on urban settings, calling attention to the number of lanes it can monitor and its mixed traffic capabilities.
FlexRadar uses a low-power 6.3GHZ radar array to detect bicycles and is also marketed for traffic signal detection. It looks like it’s meant to be permanently embedded in asphalt, like their FlexMag system for counting cars. It can differentiate between bicycles and cars, too. Their SensBike software provides a count dashboard.
They have a comparison between magnetometers and inductive loops for measuring cars that seems a bit biased.
Smartmount Bell Camera is a machine vision system that looks like it’s typically installed at intersections to capture multiple approaches at once.
FHWA Bicycle-Pedestrian Count Technology Pilot Project
Conducted in December 2016. Pedestrian counters cost anywhere from $750 to $4,000, while bicycle counters typically cost $1,650. All MPOs but one opted for portable counters.
Passive infrared counters were built by TRAFx and Eco-Counter. Inaccuracy came from users stopping in front of counters, multiple users passing by simultaneously, and diverting around the counter. The counter cannot distinguish direction. Urban counters need to point away from the road to not pick up vehicles.
Pneumatic counters from Eco-Counter and JAMAR Technologies were used to count vehicles and bicycles. The system uses axle spacing to distinguish between cars and bikes and can measure speed and direction. Damage to the tubes cause the counters to stop functioning, and the tubes are vulnerable to parking cars and vandalism.
MicroRadar sensors use radar technology to detect passing vehicles and are embedded in the pavement using an adhesive. Sensys Networks sells counters with this technology. There are still counting errors with MicroRadar.
NCHRP Methods and Technologies for Pedestrian and Bicycle Volume Data Collection: Phase 2
Published in July 2018. The National Cooperative Highway Research Program prepared this report to capture additional counting technology available as of 2018.
FHWA Coding Nonmotorized Station Location in the 2016 Traffic Monitoring Guide Format
Published in November 2016. Count stations are called “station locations” in traffic monitoring terminology. This document provides a technical reference for encoding station locations and reporting counts for those locations. The format is a fixed-width, columnar ASCII format.
Portland State University TREC
- BikePed Portal: National Bicycle and Pedestrian Count Archive
- Online Non-motorized Traffic Count Archive (2016)
- Quality Counts: Developing Accuracy Checks for Bike and Pedestrian Count Data (2019)
- Exploring Data Fusion Techniques to Derive Bicycle Volumes on a Network (2021)
UCLA Lewis Center
Bicycle and Pedestrian Data Subcommittee
A Transportation Resource Board subcommittee that supports other committees in counting methodologies for bikes and pedestrians. Their products page shows their results.
National Bicycle and Pedestrian Documentation Project
pedbikeinfo is a website run by the U.S. Department of Transportation. It looks like it was last updated around 2018. Its page on Counting and Estimating Volumes links to several other resources and provides this rationale:
They put crash data in context to better understand the exposure to risk and can be used to help estimate social, economic, and health impacts of walking and biking.
The Guide for Scalable Risk Assessment Methods for Pedestrians and Bicyclists provides a framework for estimating the exposure of bicyclists and pedestrians to harm.
Depending upon the desired geographic scale, one or more of these three analytic methods can be used to estimate pedestrian and bicyclist exposure: 1) site counts; 2) travel demand estimation models (several different types); and, 3) travel surveys.
A brief is also provided as Bicycle and Pedestrian Count Programs: Summary of Practice and Key Resources.
The brief focuses on the aspects of count programs, including site selection, equipment, and data management, with a discussion and references for more information.
- The BikePed Portal collects data from multiple cities across the United States and presents them as a dashboard.
- The Delaware Valley Regional Planning Commission’s counts map shows individual stations and a histogram of each.