How sensors and IoT are reshaping urban mobility

Urban mobility is undergoing a significant transformation. While smart cities have long explored the potential of connected systems, recent advances in sensor accuracy, deep learning, and data integration are taking urban mobility solutions to a new level.

With camera sensors, IoT, AI and digital twins all evolving, and increasingly being integrated into single platforms, city authorities, road operators, and public transport providers are beginning to change the way they plan and manage mobility services. Sensors and AI are becoming essential tools for decision-making – helping to optimise traffic lights and intersections, improve safety at tram stops, and model crowd flows more accurately during events. Their role goes far beyond monitoring – they are now shaping the design, operation, and future direction of mobility systems.

Evolving traditional mobility operations

Smart parking provides one of the clearest demonstrations of this shift. Whereas legacy systems relied on single-point sensors, connected video sensors can monitor entire zones at once, identifying not just whether a space is free, but what type of vehicle is parked there and how long it has been stationary. When integrated into a wider IoT network, this data can feed live information boards, navigation apps, or city dashboards, giving drivers and operators a real-time view of parking availability. 

The benefits are significant: reduced congestion and emissions from vehicles searching for spaces, more efficient enforcement, and dynamic pricing or allocation strategies based on actual usage. For city authorities, this translates into better resource management and higher-quality mobility data that can inform broader planning decisions.

Sensors and AI are becoming essential tools for decision-making – helping to optimise traffic lights and intersections, improve safety at tram stops, and model crowd flows more accurately

Traffic management is undergoing a similar transformation. Instead of relying on fixed, pre-set signal cycles, adaptive traffic systems powered by camera analytics can adjust light sequences dynamically according to live traffic density and flow direction. Deep learning capabilities enable the system to recognise different road users – whether pedestrians, cyclists, cars, or buses – and respond accordingly. For example, cities can prioritise public transport at key junctions during peak hours or extend pedestrian crossing times when footfall increases.

This not only helps reduce congestion and travel times but also supports safety objectives and wider sustainability goals by cutting unnecessary idling and emissions. The result is a more fluid, efficient, and human-centred mobility experience for everyone on the road.

Beyond this, and even branching beyond mobility and into safety applications, cities are increasingly using camera and IoT technologies to monitor movement during large-scale events. Whether it’s New Year’s Eve, major concerts, or political demonstrations, these systems can track the flow of people and vehicles, generate heat maps to highlight crowd concentrations, and count attendees with high accuracy. This allows them to respond quickly to safety issues, redistribute traffic, and maintain smooth operations for public transport and emergency services alike.

Expanding beyond visual data

While camera sensors are central to urban mobility management, their potential extends well beyond imagery. Modern cameras are increasingly equipped with additional sensing capabilities, including acoustic sensors that capture environmental metadata. By detecting noise levels, sudden loud sounds, or patterns of human activity, cities gain insight into their urban environment that cannot be derived from visual data alone. For instance, a road with high vehicle density may appear congested, but microphones can reveal aggressive driving behaviour, excessive exhaust noise, or crowded pedestrian areas. This richer data informs more nuanced decision-making.

Radar sensors are also playing a role in capturing complementary information like this, too. Combined with camera systems that classify objects, radar can detect vehicle speed – even in difficult weather conditions where other technology would struggle  – providing more granular insights into traffic flows. These datasets are particularly valuable when feeding into digital twins, helping cities to model how people and vehicles interact over time and to plan interventions that improve overall mobility. 

Deep learning and computer vision underpin the effectiveness of these multi-sensor ecosystems. Advances in AI have dramatically improved the accuracy of object detection and classification, reducing false positives and enabling the reliable generation of large-scale datasets. When imaging, audio, radar, and IoT devices are combined, cities obtain a comprehensive understanding of traffic patterns, environmental conditions, and public behaviour. The result is a more complete, real-time picture of urban mobility that supports smarter planning and operational decisions. 

Enhancing public safety and environmental outcomes

Applying these technologies in mobility operations does more than improve efficiency in transport networks – they also support cities’ public safety and environmental goals. Many of the same systems used to optimise traffic flow can be leveraged to reduce accidents, protect pedestrians, and cut air pollution.

A practical example comes from a tram operator in Australia, where safety at stops was a major concern. When a tram halts, signage instructs approaching vehicles to stop, preventing potential collisions with passengers disembarking. By integrating cameras with AI and radar, the system now detects approaching vehicles and prevents tram doors from opening until it is safe.

In European city centres, low-emission zones use cameras and sensors to regulate traffic flow, creating virtual gates that limit access or slow vehicle speeds

Licence plate recognition allows authorities to track repeat offenders, while strobe sirens provide immediate visual alerts to drivers. This multi-layered approach has already reduced the number of vehicles bypassing trams unsafely, demonstrating how sensor-driven systems can directly enhance public safety.

Environmental benefits are also increasingly apparent. In European city centres, low-emission zones use cameras and sensors to regulate traffic flow, creating virtual gates that limit access or slow vehicle speeds. This approach reduces air pollution, mitigates noise, and lowers the risk of accidents. By combining mobility management with environmental monitoring, cities can achieve multiple objectives simultaneously – creating safer, cleaner, and more liveable urban spaces.

Making mobility data more useful

As cities collect more mobility data, the challenge is no longer just about gathering information but about making sense of it in ways that support better planning and faster decision-making.

This is where digital twins come into play. By combining live sensor data with virtual models, cities can simulate how their mobility systems behave under different conditions and explore the impact of interventions before putting them into practice. In Australia, a major city is piloting a digital twin for event management. By integrating data from cameras and AI, the city can monitor crowd movements during significant gatherings.

The system feeds live information to control centres, police, and emergency services, allowing them to redirect traffic, maintain public transport schedules, and ensure emergency vehicles can navigate congested areas. The result is a dynamic, responsive system that enhances safety while keeping the city functioning smoothly.

In Europe, where some of these digital twin projects are a little more prevalent, Tampere in Finland has integrated traffic camera data into a digital twin to visualise flows at critical points in the city centre. In an example from Italy, a network originally designed for air quality monitoring was augmented with cameras to capture traffic patterns. This integration allows authorities to correlate emissions data with vehicle movements, supporting evidence-based policies to reduce pollution.

Cities can simulate how their mobility systems behave under different conditions and explore the impact of interventions before putting them into practice

On a more granular scale, we’ve also seen digital twins being used to compare intersection efficiency versus roundabouts. Cameras provide detailed vehicle movement data, feeding models that test lane adjustments, flow priorities, and other design modifications – all before implementing physical changes in the city.

The key advantage of digital twins lies in their predictive and modelling capabilities. Cities can simulate scenarios ranging from traffic surges at major events to infrastructure changes, allowing decision-makers to test interventions, forecast impacts, and optimise mobility strategies without the risks and costs of trial-and-error in the real world. 

Addressing challenges and exploring opportunities

Looking ahead, cities face both challenges and opportunities in expanding these technologies. AI is at the forefront, but the more processing required for AI models, the greater the demands on camera hardware. To address this, companies like Axis have developed in-house deep learning chipsets, such as the ARTPEC-9, which allow more computation to occur at the edge. For highly intensive workloads, cloud or server-based analytics remain necessary, creating a balance between local and remote processing.

This balance is ongoing. As chipsets become more capable, AI partners push their models further, and cities naturally expect more. Managing expectations and ensuring that technology is applied effectively is as important as developing new hardware. Sometimes, cameras are not the right solution, and simpler sensors or approaches may be more practical for a given challenge.

A core principle in evolving technology for cities is collaboration. By working closely with customers, technology providers can identify real challenges and build solutions that deliver tangible benefits

Innovation is also expanding beyond imagery. Acoustic sensors, radar, and LiDAR are being explored to add dimensions like noise and environmental monitoring, high-precision speed detection, and predictive modelling for incidents. These technologies feed into digital twins and support complex analytics for mobility, safety, and environmental management.

A core principle in evolving technology for cities is collaboration. By working closely with customers, technology providers can identify real challenges and build solutions that deliver tangible benefits. This might involve enhancing camera resolution and AI accuracy, deploying complementary sensors, or advising when alternative approaches are more effective. The goal is to ensure that cities are using the right mix of tools to improve mobility, safety, and environmental outcomes, rather than adopting technology for its own sake.

As cities continue to expand their digital infrastructure and embrace new technologies, the potential for innovation in urban mobility is vast. From reducing congestion and emissions to enhancing pedestrian safety and optimising public transport, the marriage of sensors, AI, and digital twins is creating smarter, safer, and cleaner urban environments.

The key lies in integrating these technologies thoughtfully – ensuring they address real-world challenges, deliver measurable benefits, and evolve alongside the changing needs of modern cities.