Case Study: Where Active Travel Meets Connected Vehicle Data — See.Sense and Compass IoT

When a council changes a street, it needs to know whether the change worked for everyone who uses it — not just drivers, and not just cyclists.

See.Sense and Compass IoT each provide a connected data source that captures how people actually move through the road network. See.Sense captures the experience of people on bikes and e-scooters. Compass IoT captures the behaviour of connected vehicles. Combined, they give road managers a multimodal view of street performance.

Our collaboration began in the iMOVE-funded Integrated Connected Data research project, where both datasets were fused with other traffic data by the University of Melbourne. In February 2026, we brought it to life for the Compass IoT Australasian National User Group, jointly presenting a before-and-after evaluation of a Dutch-style roundabout in Surf Coast Shire, Victoria — first through the eyes of cyclists, then through the eyes of drivers.

Over 140 transport professionals from national, state and local government across NSW, WA, QLD, SA, VIC, TAS and New Zealand registered for the session.

The Challenge

Traffic management has long relied on fixed sensors — loop detectors, counters and CCTV. These are valuable, but they only see what passes a fixed point, and they say little about how safe a movement felt.

Connected vehicle data closed much of that gap for drivers. But road managers told the iMOVE researchers they need data that captures all modes, including micromobility, pedestrians and active transport users.

For cycling, the gap is acute:

• Surveys and consultation are limited and subjective

• Counters and cameras cover only a handful of locations

• Fitness apps skew towards particular rider groups

• Crash data misses near misses, which are massively under-reported

The result: a street can be redesigned for cyclists with no objective way to show whether it worked for them or what it did to traffic.

The Approach: Two Connected Datasets, One View of the Street

The two datasets measure the same street from different seats and use a common language of braking, steering and speed.

Representative cycling data

At Surf Coast, the council chose participants from an oversubscribed expression-of-interest survey to match its target mix of gender, age and trip types — from school runs to commuting and leisure. Lights were offered for $10 instead of the usual $240 to avoid skewing towards cycling enthusiasts.

Before-and-after, both ways

See.Sense compared two near-identical periods either side of construction (1 Sep 2023 – 16 Jun 2024 and 1 Aug 2024 – 18 May 2025). Compass IoT ran the same comparison in its Path Analysis tool, drawing directional paths for each approach and turning movement at the roundabout.

"We look at what's normal for that rider and then score all of the manoeuvres that they take on the bike." — Philip McAleese, CEO, See.Sense

Proof Point: A Dutch-Style Roundabout That Worked for Everyone

Fischer Street in Torquay is a key north–south cycling connection between the town centre and the education and services precinct, with a history of cyclist crashes. Supported by the Transport Accident Commission (TAC) and Federal Black Spot funding, Surf Coast Shire Council retrofitted a compact Dutch-style roundabout at Inshore Drive, with extended centre islands, sharrow markings and raised pedestrian crossings. The design moves cyclists inboard, crossing alongside pedestrians and away from circulating vehicles.

What cyclists experienced (See.Sense)

• 65% reduction in swerving

• 50% reduction in harsh braking

• 41% improvement in road surface quality

• Similar cycling speeds through the junction

High-intensity events that had occurred right through the roundabout were now largely confined to the exits, where riders merge back onto the road.

What drivers experienced (Compass IoT)

• Lower 85th percentile speeds — decreases of up to 10 km/h on some approaches, e.g. northbound cars decreased from 39 km/h to 29 km/h at the midpoint

• Speed distributions decreased — The distribution of vehicles travelling at higher speeds decreased, e.g. after the change ~56% of northbound cars travelled at 40-50km/h through the roundabout

• Smoother driving — braking and steering g-force distributions shifted from higher-intensity bins and into the lowest (0–0.1g) bin, e.g. for northbound cars there was a total reduction of harsh right swerving (0.3-0.4g)

• No loss of efficiency — travel time through the roundabout stayed around 12–13 seconds

Image credit:  Compass IoT

Image credit:  Compass IoT

Read together

Cyclists rode more smoothly, drivers slowed down, and nobody's journey got longer. As Philip McAleese put it to the user group, it was a win-win-win.

Neither dataset alone could have shown this. Cycling data proves the treatment reduced risk for riders; vehicle data proves it calmed traffic without hurting network efficiency — the balance every council is trying to strike.

The Research Foundation: iMOVE Integrated Connected Data

The Surf Coast evaluation builds on a larger collaboration. In the iMOVE CRC project Integrated Connected Data for Safer, More Efficient Transport Management, led by ITS Australia with the University of Melbourne, researchers fused:

• Compass IoT connected vehicle data — including more than 41 million vehicle trajectories from 2022

• See.Sense bicycle telematics — from around 1,000 Melbourne cyclists collected between 2021 and 2024

• SCATS signal data, loop-detector counts, CAVI trajectories, crash records and network maps

The project was funded with the TAC, Transport for NSW, Transport and Main Roads Queensland, the Victorian Department of Transport and Planning, Main Roads WA and the federal infrastructure department.

What the combined data showed

• 59% of bicycle-involved crashes occurred at intersections, versus 41% mid-block, with hotspots in city centres

• Harsh braking and swerving strongly correlated with potential crash sites, giving authorities an early warning system before crashes occur

• Cyclists faced longer delays at major intersections with car-prioritised signals, and rode faster and more smoothly on dedicated bike paths

• Connected vehicle data, at roughly 2% of traffic, was enough to reveal driving styles at intersections and improve AI-driven signal control

By combining the datasets, the project brought cycling into a multimodal traffic management framework alongside connected vehicle data. The work was shortlisted for the ITS UK Better Mobility through Technology Award 2025.

"Projects like this demonstrate what's possible when connected micromobility data is included in transport big data, making it possible for multi-modal traffic management." — Irene McAleese, Co-founder, See.Sense

Compass IoT co-founder Emily Bobis described the project as showcasing "the impact that public-private partnerships have in advancing safer roads".

More Insights from Combining the Data

• Hidden micro-hazards. See.Sense data showed braking and swerving spiking on Tuesdays — bin collection day — where bins narrowed the shoulder. Surf Coast widened the shoulder buffer for tens of thousands of dollars rather than building a segregated lane. Vehicle data can confirm that changes like this don't compromise traffic flow or network efficiency.

• Smarter counter placement. Cycling route data often shows riders using a parallel street rather than the assumed corridor. It can guide where to put counters and cameras, which then calibrate the share of cyclists captured.

• Exposure-based risk. In Helsinki, combining See.Sense devices on a subset of e-scooters with fleet-wide trip data let risk be prioritised by exposure — the same principle applies when vehicle volumes are known.

• Pavement condition. See.Sense standardises road-surface readings across bike types, so councils can spot deterioration before potholes form.

Outcomes

The collaboration demonstrates that road managers can:

• Evaluate infrastructure for every road user, not just one mode

• Prove safety gains without trading off efficiency — lower speeds and fewer evasive manoeuvres, with no added travel time

• Act on leading indicators of risk instead of waiting for crashes

• Build stronger business cases for funding, using objective before-and-after evidence

• Target low-cost interventions where the data points to a specific cause

Why This Matters

Councils are constantly balancing active travel against network performance. Too often, cycling improvements are assessed only by counts, and traffic impacts only by modelling.

Combining connected cycling and connected vehicle data shows whether a street is safer for people on bikes and still works for everyone else based on real-world behaviour.

For the many Compass IoT users who already work with vehicle g-forces and speed profiles every day, See.Sense data adds the missing perspective, with a format that is directly comparable.

Key Takeaways

• Near-miss data is a leading indicator of both cyclist and driver risk

• Cycling and vehicle datasets speak the same language of braking, steering and speed, making joint analysis straightforward

• Representative deployment of cycling sensors avoids demographic and behavioural bias

• Even small, low-cost design changes can produce measurable safety improvements

• Safer cycling outcomes need not slow traffic or reduce network efficiency

Explore Combined Cycling and Vehicle Insights

See.Sense works with cities, transport agencies and road safety organisations worldwide, with 50+ projects, 120,000+ sensor-enabled lights and trackers, and more than 20 billion data points analysed.

If you already use Compass IoT and would like to add the cyclist's perspective to your next evaluation, contact Irene McAleese at irene@seesense.cc.

Sources

• Compass IoT Australasian National User Group, 17 February 2026 (recording) and Compass IoT post-event summary handout

• See.Sense presentation to the Compass IoT User Group, 17 February 2026

• ITS Australia / iMOVE: Integrated Connected Data for Safer more Efficient Transport Management (May 2025)

• See.Sense: Surf Coast Shire case study

• Compass IoT: Measuring the effectiveness of road treatments on the safety of vulnerable road users
  

Footnotes

1. "A g-force (G) is a measure of acceleration. It provides insight into how a vehicle moves - how quickly it speeds up, slows down or turns - and indicates how intense or dangerous those movements might be." (Compass IoT, Understanding Near-misses and G-forces. 2026).