Showcasing Real-World AI-Powered Radar Perception

CES 2026 will again be a key moment for the mobility and autonomy ecosystem, and Provizio’s technology will be featured in several applications, showcasing how software defined radar and AI powered perception are already delivering real world results.

As regulatory requirements and customer expectations continue to rise, perception systems must deliver more accuracy, richer environmental understanding, and greater scalability. At CES 2026, our partners will demonstrate how Provizio’s software transforms radar into a high fidelity perception platform, enabling practical, scalable solutions for today’s vehicles and tomorrow’s autonomous systems.

Reimagining Radar Through Software and AI

Radar is rapidly evolving from a hardware centric sensor into a software driven perception platform. Provizio is defining the standard, delivering the application layer of software defined radar that transforms raw sensor data into high fidelity perception and beyond.

At CES 2026, we will showcase how enhanced signal processing combined with advanced neural networks converts raw radar data into actionable perception. Visitors will see how this approach improves spatial resolution, object separation, and robustness across a wide range of operating conditions, all through software.

Enabling High-Performance, Centralised Perception

One of the most important trends shaping modern vehicle architectures is the move toward centralised processing and the software-defined vehicle. By shifting advanced AI and machine learning workloads to powerful central compute platforms, perception systems can achieve higher performance without increasing sensor complexity.

Provizio software is designed to take full advantage of this shift, enabling advanced AI and machine learning workloads to run on powerful central platforms while keeping sensor hardware efficient.

In collaboration with our partners at CES, we will highlight how this architecture delivers higher perception performance, supports continuous software evolution, and reduces total system cost, making it suitable for both current production programs and future autonomous platforms.

Visit us at CES 2026

Experience Provizio’s software defined radar performance in action at the Texas Instruments meeting rooms in LVCC North Hall, N116.

Join us to see working software, measurable performance, and real applications that demonstrate how AI powered radar perception is ready for deployment today.

We look forward to connecting in Las Vegas and sharing how Provizio is shaping the future of perception through software, AI, and scalable radar architectures.

Not only is it possible to create free space maps using radar alone, but there are also many cases where radar is superior. In this blog, we will explore how we create free space maps and how this approach performs across a number of common but challenging environments.

Our Architecture

Before freespace can be estimated, radar pointclouds must be processed and filtered. Microservices pre-process the radar data for upstream tasks in our 5D Perception® system. Each microservice involved in freespace estimation can be seen in Figure 1:

• Odometry calculates the position, velocity, orientation angle and rate of orientation angle change
• The Dynamic Target Filter removes dynamic targets so that only the static scene remains
• SLAM accumulation increases pointcloud density by overlapping past pointclouds with new pointclouds

The freespace estimation algorithm uses the pre-processed pointcloud to create an accurate freespace map. Figure 2 outlines the individual steps of the algorithm. The goal of our freespace estimation algorithm is to identify freespace beyond what can be seen by ray tracing. A generalised approach to freespace estimation allows us to incorporate multiple detections along each beam. This makes it possible to estimate freespace that might be out of direct line-of-sight but that is still visible to the radar.

Performance in Challenging Environments

Curved Roads and Complex Junctions

Freespace estimation algorithms like Inverse Sensor Modelling (ISM) struggle to estimate freespace around corners. This is largely due to the data losses incurred when 3D pointcloud data is converted into a 2D Bird’s Eye View (BEV) representation, which can obstruct detail at further ranges. Figure 3 shows a road scenario where ISM has reduced range. Our freespace estimate follows the curvature of the road beyond line-of-sight to 50m+.

Obstacles at Long Range

Detection range is important for fast moving vehicles like cars and drones because large distances can be covered in a small amount of time. Figure 4 shows a vehicle that has stopped in the hard shoulder on a motorway. Using our sensors, this vehicle is first detected at 300m and an accurate freespace map is created at 125m. The speed limit of the vehicle is 100km/h or 27.8m/s. Therefore the vehicle has 10.8 seconds from first detecting the object to process the obstacle and 4.5 seconds to avoid the vehicle using the freespace map.

Figure 5 shows the camera and freespace estimate overlaid. The obstacle is closer to make it easier to see the freespace. Freespace narrows as expected beside the obstacle and the truck and is precisely removed.

Indoor Car Park

Indoor environments can be challenging to map because they are often dimly lit, cannot avail of GPS and have targets at close range and long range. Figure 6 shows an indoor car park environment which has harsh lighting and clutter above the radar from the car park ceiling. Detections from the ceiling are filtered out correctly and do not block the freespace estimate. Using our technique, spaces between vehicles on the left and a junction to turn right are both detected at greater than 20m.

Conclusion

It is clear that our new class of MIMSO® imaging radars are more than adequate for freespace estimation. Estimate accuracy and range is improved by looking beyond line-of-sight. Many freespace algorithms fail around curved and complex road structures but our estimate does not degrade in these conditions.

We have shown how our radar-only freespace estimation algorithm performs indoors and outdoors at a variety of ranges. Furthermore, lighting and weather conditions do not affect performance unlike LiDAR and camera systems. The low cost of the radar and processing stack makes radar-only freespace a viable option for industries such as automotive, mining and agriculture.

To learn more, read our white paper.

Imagine a vehicle that can constantly learn and improve over time, just like we drivers do. That’s the future of perception technology and ubiquitous autonomy, and Provizio is at the forefront of this exciting revolution.

Traditionally, radar systems were limited by the hardware they were built with, but with the rise of software-defined radars, we now have the ability to adapt and improve radar performance through fine-tuning and over-the-air (OTA) updates.

At CES 2024, we leveraged these technologies to continually optimise the performance of our 5D Perception® system using the unique driving environment that is Las Vegas. In the short couple of days we were demonstrating our latest technology developments to our current and future partners, our perception performance was continually learning, adapting and improving. How? Let’s take a look.

A closer look at our technology

5D Perception^®

5D Perception® refers to the unique capability of Provizio radar sensors to “see” and perceive the world in five dimensions:

• Third dimension: High-resolution 3D world view
• Fourth dimension: Relative velocity & distance of all objects in the vehicle’s environment
• Fifth dimension: Edge Intelligence: making sense of everything in the vehicle’s environment and determining a safe driving path utilising Artificial Intelligence (AI), Machine Learning (ML) and Digital Signal Processing (DSP) to deliver object classification & drivable freespace mapping.

When speaking of optimising our 5D Perception® system at CES, our main focus was to improve performance with respect to object detection, classification, tracking and freespace mapping – the 5th dimension of our perception system. To do this, we needed to re-train our Convolutional Neural Net (CNN) algorithms based on data recorded from Las Vegas roads. The catch? We had to do it LIVE!

Fine-Tuning our Neural Networks

Think of our 5D Perception® system like a student. It needs to be able to learn and adapt to different situations (generalisation), but it also needs to be good at specific tasks (fine-tuning). For example, our radars are trained on a massive dataset of roads from Europe. But the US is a different environment, with wider roads, different markings, and even unfamiliar vehicles like large semi-trucks or pickup trucks.

To address this, we continued to fine-tune our system using data captured directly in Las Vegas while demonstrating to our customers. Using our custom automated data pipeline, data collected from the vehicle was uploaded directly to the cloud. From here, the pipeline extracted and processed the data, making it available to use for fine-tuning our neural network models.

With this data, our engineers created a new dataset by combining a mix of data captured from Las Vegas with data from our general balanced dataset. Our engineers then used this data to fine-tune our existing generalised model overnight, seamlessly delivering improved performance for the next day of captures and demos.

The result: In addition to providing 35% less false-negative vehicle detections, our fine-tuning of the neural networks improved vehicle detection range by over 20m on Las Vegas streets. In addition, we also noted a 45% improvement in pedestrian tracking stability and a 25% improvement in stability for the tracking of cars and large-vehicles.

Optimising Freespace

The Freespace microservice acts as a virtual guide, identifying safe driving paths and highlighting potential obstacles on the road. Since the service relies primarily on data from point-cloud clustering, road boundary estimation and odometry, the optimisation process at CES consisted of finding the perfect balance between raising the clustering threshold while maintaining reliable road boundary estimation.

As a result of these optimisations, the freespace detection distance for Las-Vegas boulevards and large interstate roads went from an average maximum range of 53m to an optimised maximum range of 86m. This is a 62% increase in average maximum range for the detection of freespace – day and night, in all weather conditions.

Conclusion

By leveraging the power of our software-defined radars and OTA updates, we are able to perform near real-time optimisation of our perception system to provide higher performance in a variety of areas – from classification & tracking, to radar-based odometry and freespace detection.

In just a few days, our system demonstrated:

• The ability to detect vehicles up to 20m further away and with 35% less false negatives than before.
• An improvement of 45% in pedestrian tracking stability and 25% for cars and large-vehicles.
• An improvement in average maximum freespace detection range from 53m to 86m – an increase of 62%.

This is just a glimpse of the future of radar-based perception technology, where software systems working on-the-edge are constantly learning and adapting. We believe that making the transition to scalable L3+ a reality is a challenge that can be solved by continuously learning and improving our 5D Perception® solution.

To learn more, read our white paper.

The Provizio AI Approach

At Provizio, we are leveraging AI to deliver on three core goals:

1. To deliver LiDAR-level resolution performance, while leveraging the robust sensing and cost advantages of radar technology.
2. To deliver powerful on-the-edge perception capabilities while minimising resource demands.
3. To deliver a solution that can improve over time, leveraging crowdsourced datasets to improve perception capabilities and deliver enhanced Software Over The Air (SOTA) features to our customers.

Our Implementation

With 5D Perception, Provizio AI is utilised in a unique Tri-Level design to deliver compound enhancement from the signal level, through to the point cloud, and finally at the fusion level. In this way, we make the most of our hardware systems by using intelligent software to squeeze out the best possible performance from each layer of the stack. Let’s start with the first layer – point cloud denoising.

Neural Networks for Point Cloud Denoising and Enhancement

Stage 1: Training Dataset

The first phase in creating an effective AI model starts with good quality training data. At Provizio, we generate such data using a 3-stage process:

1. Sensor Synchronisation: Inputs from radar, LiDAR & camera are used to create an accurate source of ground-truth data. Inputs from each respective sensor need to be synchronised, such that objects and their locations are consistent.
2. Point Cloud Filtering Based on Lidar Data and Camera Semantic Segmentation: During this process, LiDAR and camera data is compared. Using semantic segmentation, each individual pixel from a video frame is classified into categories like “road”, “pedestrian”, “vehicle” and so on. This is cross referenced with 3D LiDAR point cloud data to ensure any “noise” (erroneous points on the point cloud) are removed.
3. Manual Corrections: Human visual inspection is used to ensure the resultant training data is as accurate as possible.

Stage 2: Lightweight Models for Real-Time denoising

2D Convolutional Neural Networks (CNN): With this method, 3D radar point cloud data is transformed into 2D projections, which simplifies the training process. A unique CNN architecture is then trained on a dataset where noise in the radar data has been identified and labelled. During this process, the CNN learns to identify patterns in the data that represent noise and as a result, once training is complete, the CNN can be used to identify and filter noise from previously unseen data.

Stage 3: Point Cloud Super Resolution

In a similar way to how noise patterns are identified and removed using CNNs, patterns that denote real objects can also used to improve the resolution of point cloud outputs. In this case, during the training process, the CNN learns to understand the spatial relationships within high-resolution ground truth point cloud datasets and predict where additional points should be added to increase resolution. Once trained, the CNN can take lower resolution point clouds and enhance them by adding additional points in a way that increases detail and accuracy.

Neural Networks for Object Detection, Tracking and Freespace Estimation

Once the data from our radar sensors is de-noised and enhanced as per the above systems, a further set of neural networks is used to process this data with the goal of understanding the real-world environment it represents. In this respect, our hardware and software teams worked closely together to develop an understanding of how to build a neural network that could extract the most information from the radar point clouds. In doing so, several efficiencies in the process were identified to create a lightweight system, capable of performing advanced perception tasks on-the-edge.

The Provizio Advantage

The above provides a high-level outline of the modular process we use in Provizio to maximise the value output of our products. Not only does this approach enable greater maintainability over time, but by developing both the hardware and software for our devices, we posses a unique ability to produce high quality outputs at a fraction of the cost of our competitors. By leveraging AI within our 5D Perception system, we deliver:

• Robust reliability in challenging environmental conditions.
• Native support for a triple redundancy safety approach, incorporating VizioPrime, Plex and camera sensors.
• Real-time, on-the-edge processing that slashes hazard response times and reduces manufacturing cost and complexity.
• Scalable, over-the-air improvements that deliver improved safety and unique features throughout the lifetime of the product.

Conclusion

The application of AI in vehicle perception is a field rich with innovation and challenges. As AI continues to evolve, Provizio is at the forefront of addressing the complex technical hurdles affecting the safety and real-world variability of autonomous systems, such that a future of zero accidents will become possible for all.

The Necessity of Accurate Localisation and Mapping

For an autonomous vehicle, understanding its precise location in the world and having a map of its surroundings is crucial, as without it, the vehicle would have no ability to navigate to a destination. However, in the context of most modern AVs, the level of mapping and localisation accuracy required is significantly greater than what you may expect from Google Maps on your smartphone.

Localisation

Localisation refers to the vehicle’s ability to determine its position relative to a map. Unlike traditional GPS systems that offer accuracy within meters, AVs require centimetre-level precision to ensure they don’t stray outside of their lane or take the wrong turn in a complex driving environment.

Mapping

While a typical map shows routes, lanes and sometimes buildings, maps for AVs typically require much more detailed information, including road layouts, traffic signs, lane markers, and other critical infrastructure. Such maps are often referred to as High Definition or HD maps.

Current Technologies in AV Localisation and Mapping

Real-Time Kinematic (RTK) and Differential GNSS (DGNSS)

Overview:

Standard GNSS (Global Navigation Satellite System) signals can have errors due to atmospheric conditions, satellite clock inaccuracies, and other factors, resulting in positional inaccuracies. To correct these inaccuracies, ground-based stations known as base stations are used. These stations are at known, fixed locations and continuously monitor GNSS signals to calculate the difference between its known precise location and the location indicated by the GNSS signals. This data is used to generate correction information, which is incorporated in RTK or DGNSS receivers to enhance localisation accuracy.

Challenges:

While RTK and DGNSS can provide centimetre-level localisation accuracy, they suffer from a variety of issues:

1. Infrastructure Dependency: The effectiveness of such systems depends on the availability, proximity & maintenance of base station networks, which can be a constraint in terms of scalability and coverage.
2. Latency in Correction Data: There can be a delay in receiving correction data, which can affect the accuracy in highly dynamic driving scenarios.
3. Reliability in Diverse Environments: RTK & DGNSS systems often struggle in urban environments or enclosed spaces (such as multi-story or underground car parks) where buildings and other objects can obstruct or interfere with received signals. In such cases, localisation accuracy is often significantly reduced.
4. Cost: RTK & DGNSS systems require specialised equipment and are considerably more expensive than standard GPS/GNSS units.

HD Maps

Overview:

High Definition (HD) maps contain comprehensive, highly accurate data about roadways and their surroundings, including:

1. Lane Information: Exact lane positions, widths, and the direction of travel.
2. Traffic Signs and Signals: Locations and details of traffic signs, signals, and road markings.
3. 3D Environmental Data: Information about buildings, trees, curbs, and other static objects around the road.
4. Road Geometry: Precise details about the road’s curvature, slopes, and elevation changes.

In AVs, HD Maps are used to provide centimetre-level localisation accuracy, even in the absence of a GPS signal. This is done by matching data from onboard sensors, such as LiDAR and camera, with the information stored in the HD map to determine the vehicle’s precise location.

Challenges:

While HD Maps overcome the issue of providing centimetre-level localisation accuracy in environments without a GPS signal, they also suffer from a variety of issues:

1. Maintaining Up-to-Date Maps: The dynamic nature of road environments means that HD maps need to be regularly updated to reflect current conditions, such as changes due to construction or road works.
2. Data Processing and Storage: The detailed information contained in HD maps requires significant data processing capabilities and storage, both on the vehicle and in the cloud.
3. Integration with Sensor Data: Effectively integrating and cross-referencing HD map data with real-time sensor data is a complex task that requires sophisticated algorithms and computing power.
4. Scalability, Coverage & Cost: Creating and maintaining HD maps for extensive road networks is a massive undertaking, requiring substantial resources and coordination.

Introducing VizioSLAM

VizioSLAM (Simultaneous Localisation & Mapping) enables centimetre-level localisation and HD map generation, with minimal hardware complexity and cost. Let’s see how it works:

1. Leveraging the high spatial resolution of Provizio’s 5D Perception sensors, novel software techniques are used to extract visual reference points from radar point cloud data.
2. As the vehicle moves within its environment, this data is extracted from each new point cloud frame.
3. With every new frame produced, specialised algorithms compare the point clouds frame by frame and use the visual reference points as a means to calculate changes in the velocity and rotation of the vehicle relative to its surrounding environment.
4. With our high-speed, on-the-edge perception process, VizioPrime and VizioPlex sensors can feed these algorithms with enough data of sufficiently high resolution to provide centimetre-level localisation accuracy, without the need for HD maps or GPS.

The Provizio Advantage

1. Reliability in Diverse Environments: VizioSLAM is not negatively affected by complex environments such as dense urban areas or enclosed spaces. As a result, it is a more robust solution than RTK or DGNSS.
2. Scalability: VizioSLAM works on-vehicle and in real-time, without the need for external correction signals or HD maps. As a result, the system can be applied in any environment, without any prerequisites for HD Maps or base station networks.
3. Cost: Without the need for costly DGNSS sensors or to maintain up-to-date HD maps, VizioSLAM is a once-off cost for OEMs.
4. Integration: With our unique on-the-edge perception process, VizioSLAM can be easily integrated as part of a larger AV stack, with minimal overhead in terms of compute or network demands.
5. Mapping: Like LiDAR, VizioPrime & Plex sensors can be used to generate high fidelity environmental maps. As a result, OEMs can use our sensors to build their own crowdsourced mapping solutions, and reduce reliance on third-party HD map providers.

Conclusion

Accurate localisation and mapping are vital components in the development and operation of autonomous vehicles. While current technologies like DGNSS, IMU & HD maps have provided an initial starting point for the development of AVs, the challenges in scaling these technologies limit their appeal for mass market applications.

With VizioSLAM, we believe we have developed a compelling alternative for both accurate localisation and mapping, which can cater for the needs of mass market adoption. If you’re interested in learning more about VizioSLAM, visit the VizioSLAM webpage, where you can find more detailed technical information and demonstration videos. Alternatively, if you’d like to get in touch with us directly, please visit Contact Us.

Taking a Step Back

As a startup organisation, it’s easy to fall into the trap of following “the industry standard” when establishing the foundations of a workflow management process. With so many things to build and so little hands to build them, startups often find themselves so entrenched in the engineering process that they fail to see the bigger picture of why.

At Provizio, we understood that in order to supplant industry leading competitors, we needed to focus as much on innovation within our development process as in our products. In this regard, an “if it works for them it can work for us” approach was not going to provide the momentum we needed to ultimately build a product that could entice customers away from the grip of trusted suppliers. Hence, a key approach to continued innovation & productivity within Provizio, is the ability for us to regularly step out of the trenches and take a candid look at whether our process and priorities are truly aligned and in service of our overall mission.

The Intersection of Workplace Psychology and Productivity:

In today’s rapidly evolving business landscape, where factors like remote work, reduced hour work weeks, flexible working hours and artificial intelligence are hot topics among employees, many companies face the challenge either of holding their ground with tried and true processes, or taking a chance to embrace the unknown in the hopes of coming out on top.

At Provizio, we believe our workforce is the heart of our success and as such, we spend a considerable amount of time in pioneering innovative work practices to boost productivity, creativity, and employee satisfaction to ensure that our heart continues to beat strong and true. These practices, deeply rooted in the principles of workplace psychology, are reshaping how we think about work, collaboration, and productivity.

Agile Development

Agile principles are the bedrock of our project management approach, allowing us to respond swiftly and effectively to the ever-evolving demands of our industry. However, while many organisations get bogged down in agile process documentation, arguing between the pros and cons of the now countless agile frameworks available today, at Provizio we take a different approach.

Our work practices are guided mainly by values outlined in the original Agile Manifesto, whose deliberately vague guidelines left the door open for organisations to customise a workflow approach that worked best for their unique structure, products and workforce. In this respect, we made our own interpretations of the values of the Agile Manifesto and built a framework tailored for our specific needs.

When it comes to agile frameworks, Provizio takes a little inspiration from almost every one.

• SAFe inspired our approach for establishing yearly objectives and long-term goals that could easily be referenced and tracked throughout the development process.
• Scrum inspired our 2-week development increments, with regular touch-points for teams to align on progress and priorities.
• FAST Agile inspired our “collective” engineering structure, where engineers were given maximum autonomy to work in self-structured teams to address the dynamic needs and phases of complex product development.
• Kanban inspired our day to day workflow management system, where everyone can quickly understand the individual tasks and priorities of team members.
• Various Liberating Structures inspired our approach to meetings, where frequent and disruptive meeting schedules were replaced by more infrequent but valuable events, where everyone in the organisation could get a better grasp of our priorities and our progress.

These, and many other frameworks like LeSS, Nexus, The Spotify Model and more, all fed into the development of our current agile process which, like our products, improves and evolves incrementally based on regular feedback.

Overcoming Resistance

A frequent barrier to agile development, often voiced by those exposed to more traditional or waterfall based development styles, is the alleged lack of evidence to support the view that agile will work in their specific industry. Now, while there have been many studies to support that agile values such as individual autonomy and incremental development in the workplace have been linked to higher job satisfaction, improved productivity, increased creativity & enhanced product quality, there will always be a sense of skepticism when adopting a new process.

However, the key to overcoming such resistance is to ensure that employees have a sense of choice and initiative. In agile, nothing is fixed and scrutiny is welcome to provide the insights needed to improve and adapt the process to suit every unique case. In giving people an outlet to express their feedback over time, with the knowledge that their feedback is heard and actioned, this often leads to higher levels of intrinsic motivation, which is a key driver in producing higher quality work.

Conclusion

At Provizio, our approach to work mirrors our approach to technology – innovative, agile, and always focused on the human element. By embracing agile principles and granting autonomy to our employees, we’re not just building advanced technologies; we’re shaping a work culture that is as progressive and cutting-edge as our products.

Together, these practices form the cornerstone of our success, driving us towards our ultimate goal – making roads safer for everyone.

What is 4D Radar?

4D radar is an evolution from traditional radar systems, adding the dimension of elevation to the existing coordinates of range, azimuth, and velocity. This additional data dimension allows for more precise tracking and identification of objects in three-dimensional space and their speed.

Key Components of 4D Radar Systems

1. Transmitter and Receiver Arrays: Modern 4D radar systems typically use phased array antennas, which consist of numerous small antenna elements. Each element can be controlled independently, allowing for electronic beam steering without moving the antenna mechanically.
2. Frequency Modulated Continuous Wave (FMCW) Technology: Many 4D radar systems utilise FMCW technology, where the radar signal’s frequency is varied over time. This allows for the simultaneous measurement of range and velocity.
3. Multiple Input Multiple Output (MIMO) Techniques: MIMO technology is used to improve the resolution and accuracy of a radar system. By transmitting multiple signals and receiving their echoes through multiple channels, MIMO enhances target detection capabilities and enables improved spatial and angular resolution.
4. Fast Fourier Transform (FFT): FFTs are employed to convert time-domain radar signals into the frequency domain. This transformation is essential for determining the range and velocity of targets.
5. Doppler Processing: By analysing the Doppler shift in the frequency of returned signals, the speed and direction of an object can be determined.
6. Angle of Arrival (AoA) Estimation: Sophisticated DSP algorithms estimate the AoA of radar signals, which provides information on the object’s azimuth (horizontal location) and elevation.

Challenges with Existing 4D Radar Technology

1. Computational Complexity: The high volume of data and the complexity of the algorithms require significant processing power.
2. Real-Time Processing Requirements: The requirement for real-time signal processing for applications like autonomous driving and collision avoidance is technically demanding. There is a constant need for faster and more efficient processing algorithms.
3. Antenna Design: The complexity of designing phased array antennas for 4D radar systems is significant. These antennas must accurately manage beam steering in three dimensions, which can be technically challenging and expensive.
4. Size and Integration: Incorporating 4D radar systems into smaller platforms, such as those required for mass market automotive or mixed mobility applications, can be constrained by the size and power requirements of the hardware.
5. Manufacturing Costs: The cost of producing sophisticated phased array antennas and other components of 4D radar systems can be high. This impacts the scalability and adoption rate in various industries.
6. Complex Algorithm Development: Developing algorithms that can effectively interpret the complex data from 4D radar systems is challenging. These algorithms must be robust, accurate, and able to differentiate between a wide range of target types.

Taking 4D to the Next Dimension

Key to 5D Perception® is our advancements in radar resolution and range. Digital radars in vehicles today generally have a single radar chip, with 3 transmit channels and 4 receive channels that allow them to deliver a virtual antenna array of 12 elements. In radar, the number of elements is like the pixel count for cameras and larger virtual apertures are preferred to improve resolution. Today’s radar has a typical detection range from 100m to 150m. These sensors enable the basic and dependable adaptive cruise control (ACC) and Automatic Emergency Breaking (AEB) on vehicles. However, to enable safe autonomy Provizio brought learning from advanced aerospace applications and developed a dedicated cost-effective solution for the automotive vehicle market.

Previous to Provizio our team built products that NASA described as “awesome” and the world’s leading autonomous driving group described as “the gold standard” in automotive Radar. That experience led us to envision a completely different, patent protected (7 at the time of writing), approach to software defined imaging radar.

MIMSO®

MIMSO^® (Multiple Input Multiple Sparse Output) is a software defined active antenna technique, which embeds proprietary surface-mount technology (SMT) integrated circuits (ICs) into a novel planar antenna design. This allows us to lower the receive path noise floor and discriminate more of the radar beam by essentially recycling parts of the radar beams that have traditionally been filtered. On the transmit path, our SMT ICs allow us to carry out instantaneous beam switching, which multiplies our transmit channels and further increases our resolution. Ultimately, MIMSO allows us to extract more than 30x the resolution from each physical radar channel.

SPTDMA

Our proprietary software modulation technique, SPTDMA, is a new form of multiplexing that enables unparalleled 6K resolution out to over 600m in all weather conditions, with the added benefit of protecting against interference from other sensors.

Industry Limitations

In radar systems, there are different modulation schemes that can be used to transmit and receive signals from multiple targets. One such scheme is Time Division Multiple Access (TDMA), where each transmitter transmits a signal on separate time slots. This allows the received signals from each transmitter to be easily separated with minimal computational overhead. However, using TDMA modulation comes at a cost of reduced maximum unambiguous velocity measurement, since the frequency at which measurements can be performed is limited by the time delay introduced by the allocation of time slots for each transmitter.

To address this issue, another modulation scheme called Doppler Division Multiple Access (DDMA) can be used. In this scheme, each transmitter transmits simultaneously (eliminating the time delay) and a series of algorithms are used to recover the velocity of a detected object by correctly matching each received signal to the transmitter that sent it. However, this approach is limited by challenges in accurately determining the velocity of targets and resolution limitations for objects that are closely spaced. Moreover, the addition of complex algorithms to match transmitter/receiver signals requires sophisticated signal processing that can lead to increased computational demand and pose challenges for real-time processing.

The Benefits of SPTDMA

Provizio’s patented SPTDMA (Slow Phase Time Division Multiple Access) solution combines TDMA and DDMA techniques by splitting multiple transmitters into smaller sub-arrays, with each sub-array transmitting simultaneously on an allocated time slot. In this way, while there is a limitation on maximum unambiguous velocity compared to a pure DDMA system, we can increase the number transmit channels to improve angular accuracy and range, while keeping computational complexity low enough to enable real-time, on-the-edge processing.

The Competition

The incumbent Tier 1 radar manufacturers are all developing 4D radar to enable Level 3 driving. They have very little differentiation and are almost all building forward facing 4D radars with 192 virtual antenna elements, delivered using 4x COTS radar chips. A big improvement, but nowhere near AV requirements.

There are also other competitors working to develop super-resolution radar. These solutions generally fall into one of two camps:

• Filled Array Radars: A brute force approach where more and more physical radar channels are added until you get to 1° angular resolution. Several companies have announced they are building radars with 2304 virtual elements by using 2304 physical radar channels. This has 2 major disadvantages:
  • Cost (12x channels = 12x cost).
  • Noise (Radar has a ‘cross-talk’ problem and the proximity of radar channels exacerbates it).
• Very Sparse MIMO Radars: The homeopathic approach where you dilute the number of radar channels using a very sparse array (some across the entire vehicle!) or by using AI neural nets (NN) to infer more resolution. The problem here becomes trust, can automakers trust these sensors not to miss detections due to over dilution in the array?

The Provizio Advantage

• MIMSO® can be applied with any Commercial off the Shelf (COTS) radar chips, which allows us to licence to multiple partners and applications to deliver high resolution perception at a significantly larger scale and lower cost than previously possible.
• MIMSO® allows us to deliver 3x more resolution and 2x the range of the most ambitious filled array approaches, but at 12x lower cost.
• Provizio delivers more than 30x the resolution and double the range of the incumbent Tier 1 sensors at a similar cost to the solutions they are employing today.
• Provizio’s 5D Perception® Radar integration costs are much lower than other sensors. The MIMSO® Radar cover (the radome), is made from low-cost polycarbonate plastic i.e. the same as the car bumper, so it can be seamlessly integrated in the vehicle body. They are super robust, and bugs, dust, weather and interference have negligible impact on performance. As a result, it is possible to have 5D Perception® radar integration which does not impact the vehicle aesthetics and still delivers 600m radar performance.

Conclusion

Provizio has invested significant resources in pushing the bounds of what’s possible with 4D radar hardware and software technology. With our patented MIMSO® and SPTDMA systems powering 5D Perception®, Provizio offers finer resolution, faster processing speeds, and more accurate target detection and tracking, delivering compelling solutions to enable the transition to L3+ autonomy and the future of automotive safety.

Autonomous vehicles were meant to end road crashes while simultaneously providing low cost mobility, but this goal depended on ubiquitous autonomy which we are no closer to today than 10 years and $100bn’s of investment ago. Autonomous Vehicle (AV) groups have made impressive strides in delivering driverless vehicles in geo-fenced regions. However, while this has proved that it is indeed possible to deliver safe autonomy on our roads, the cost associated with these platforms means they cannot be mass deployed and so cannot impact the driving problem in the near term.

Provizio imagined an affordable perception system that could give every driver AV grade perception, 360° insights in all weather conditions and pave a true path to ubiquitous autonomy; that’s why we created 5D Perception®.

The Perception Problem

AV’s today use a combination of sensors and compute resources to perceive the world and take the human driver out of the loop.

• To allow an AV to see the world in 3D, the industry has developed automotive LiDAR (the large siren type devices on the roof AV’s).
• Radar is deployed all around the vehicle because of its unique doppler properties that deliver the precise range and velocity of objects.
• Cameras deliver semantic information such as the ability to read road markings and traffic lights.
• Satellites provide location data and, when combined with high-definition mapping, allows the vehicle to navigate.

These sensors are then fused on a central compute where a lot of heavy lifting software takes over, deploying processing algorithms, machine learning (ML) and artificial intelligence (AI) to deliver a passenger safely from A to B.

Unfortunately, there are several disadvantages to this approach:

1. It’s a complex and expensive platform. LiDAR is largely what allows these vehicles to succeed but as a sensor, it is expensive to build and hard to scale.
2. LiDAR needs to be seen to see and therefore impacts vehicle aesthetics. A number of companies are making inroads on building scalable solid-state LiDAR but most acknowledge and focus on forward facing perception. The balance of the system is also high cost as it requires external mounting and cleaning.
3. LiDAR struggles to perform in bad weather. Radar doesn’t have these cost or environmental issues which is why it is deployed in almost all vehicles on the road today but today’s Radars lack the resolution a LiDAR offers…until now.
4. LiDAR produces an immense amount of data which places a huge demand on vehicle compute and networking systems in order to process the data in real-time. Automakers are struggling to both reduce manufacturing and materials costs, while simultaneously increasing computational power and network bandwidth.

Introducing 5D Perception®

5D Perception is a sensor to perception level solution designed to target the core constraints impeding the progression of today’s automotive perception technology. Our proprietary super-resolution imaging radar, which sees the world in a LiDAR like 3D point cloud and also delivers the precise range and velocity of every point, provides the 4-dimensional sensing element of our 5D solution. Uniquely, the entire radar backend is built on a Graphics Processing Unit (GPU), which allows us to deliver AI perception on-the-edge. This is what we call the 5th D in our 5D Perception® solution. This offers multiple advantages to our partners and ultimately all road users:

• Edge perception means edge decision making leading to driving interventions which are 30–100x quicker than today’s vehicles. Reduced intervention time = increased safety.
• The system allows for low level fusion with other sensors such as camera and LiDAR, enabling unique insights and redundancy.
• By running ML and AI Neural Nets on the edge, data from the sensor can provide high-level information such as object detections, localisation information and hazard warnings, without the need to send data hungry point clouds through the vehicle network. In doing so, the cost of on-board vehicle compute and networking can be significantly reduced.
• Since the entire sensor is software defined, Software Over The Air (SOTA) can be leveraged to deliver enhanced customer functionality, without the need to purchase new hardware. This brings an additional level of flexibility to consumers, while minimising manufacturing complexity.
• Leveraging our proprietary super-resolution imaging radars, coupled with our super-resolution AI enhanced point clouds, 5D Perception can enable accurate Simultaneous Localisation & Mapping (SLAM) without the need for GPS. This resolves many of the performance constraints of existing sensors when used within tunnels or underground car parks.

An Accelerated Path to Production

To deliver 5D Perception® and get the most from our high-resolution radar, we developed a perception stack which is designed to leverage best-in-class sensors from partner suppliers of camera and LiDAR sensors. This provides a clear path for OEM and AV partners to develop increasingly complex safety and autonomous solutions including:

• Level 3/4 Safe Augmented Driving: Crash resistant driving and limited autonomy can be realised with 360° high resolution radar and camera coverage, coupled with best-in-class LiDAR acting as the primary forward facing perception sensor.
• Level 4 Safe Autonomous Driving: By beefing up the forward-facing radar range and resolution we now have two super-resolution forward facing perception sensors which will enable extended autonomy.
• Level 5 Safe Autonomous Driving: Using fused super-resolution radar and camera sensing in full 360° to create a 1km cocoon of perception unlocks ubiquitous autonomy. In this configuration, LiDAR can be deployed as a redundant safety sensor on partner’s L5 platforms.

Teamwork to Make the Dreamwork

Since foundation we promised to partner, partner, partner to achieve our goal of zero road deaths and a true path to autonomy. The real key to this technology achieving mass adoption is the ability for it to be implemented at scale, with Provizio working alongside OEM’s and Tier 1 auto manufacturers to bring this technology to series production and on our roads by 2025.

Our core IP licence model has us working with some very exciting partners already and we will be announcing additional partners throughout 2023, however we are always open to more. We are looking for:

• Automotive Tier 1s with series production opportunities who can licence MIMSO® to reduce production cost or ramp up resolution — this allows partners to supercharge their Radar solutions.
• OEMs can licence our perception stack to fill out their driving experience — our goal is to provide the full perception stack to our customers to reduce their time to market.
• We are partnering with leading LiDAR and Camera providers to allow them to deliver improved fused solutions while simultaneously lowering their overall unit costs — this provides a win win partnership to leverage individual partner key strengths.
• We are partnering with central processing partners to fully realise our 5D Perception® goals — we recognise that 5D Perception® requires computation power and working with partners who can provide this decreases our customer adoption time.

By leveraging our proprietary 5D Perception® technology, Provizio can supply class leading systems for a fraction of the current cost and with low integration complexity. That has led to applications beyond traditional automotive. On any one-day, Provizio and our partners are testing on roads in Shannon and Palo Alto, in German cornfields, down Pittsburgh mines or scooting in Stockholm. We’d love to hear about your challenges and how 5D Perception® and MIMSO® could make an impact.

Our Initiatives

The 4-Day Work Week:

Provizio is a firm believer in “it’s not about how many hours you work, it’s about what you do with the hours that you do work”. While the logic of this sentiment is something many people agree on, the concept of a 4-day work week is one which still scares many employers. Despite various research studies highlighting the potential benefits of a 4-day week, the thought of losing eight hours of precious development time each week proves too difficult a pill to swallow for many organisations. However, at Provizio, like with anything we do in our organisation, we approached the 4-day week like an experiment – create a hypothesis, test the hypothesis and evaluate the results.

In adopting a 4-day work week, a critical enabling component was to first understand how our existing 5-day process would adapt to ensure productivity wasn’t compromised. This included evaluating meeting schedules, understanding common productivity disruptors, assessing our team alignment processes, and ensuring that individuals were empowered to enhance their individual day to day productivity.

After a successful trial, it was amazing to see just how much our workflow process changed in accommodating a reduced hour workweek. New initiatives like “Deep Dive Days”, meeting blocks and asynchronous chat and video updates enabled vast reductions in daily disruptors, while an extra day of rest each week enabled employees to spend more time with family, friends, or on their favourite hobbies, leading to increased productivity, job satisfaction and work life balance.

Remote Work and Flexible Hours:

Remote work and flexible working hours are other critical aspects of Provizio’s work life balance approach, aligning with current workplace psychology research. One study found that employees working remotely reported higher job satisfaction and reduced attrition rates, owing to employees being able to create a personal work environment conducive to their productivity. Flexible working hours also play a significant role, as evidenced by a 2016 study which found that flexible schedules led to reduced stress and burnout.

In our own experience, remote work and flexible working hours are a critical enabler of enhanced creativity and productivity for most employees. Our internal studies showed that many employees use both flexible working hours and remote work to their advantage, taking time during the traditional workday for running errands or minding children, and using quiet time later in the evening for deep focus.

In addition, many people have different circadian rhythms, with some preferring to do their most complex work in the morning and others preferring to work at night. With a flexible and remote work approach, employees are free to utilise the work day to their advantage, maximising productivity while minimising distraction. This approach isn’t just good for our employees; it’s beneficial for our business. A rested and fulfilled team brings more energy and creativity to their work, driving innovation and helping us stay ahead of competition.

Productivity Boosters:

In addition to the initiatives outlined above, we found the below techniques to be very helpful in improving overall productivity. However, it is key to remember that not every approach is suited to every person or role, so please take the time to assess what may work best for your unique situation.

• Time-box your work in regular focus intervals, separated by short breaks (Pomodoro Technique). This often helps to reduce distractions & improve consistency in your daily workflow.
• Create a quick plan of your work at the start of each day, with set timeframes within which you expect to complete your tasks. This can give you a stronger intention to complete your work than if you just go about completing tasks in an ad-hoc manner.
• Ensure any meetings you create or attend are valuable, have an agenda, are time-boxed, have action items and are attended by a relevant audience.
• Use meeting-free deep focus time to enable complex work to be delivered without distractions.
• Use online whiteboarding tools and meeting timers to keep everyone engaged, to record action items in real-time, and to maintain a focus on accurate timekeeping.
• Try and group meetings together, where possible, to avoid ad-hoc distractions throughout the working day.
• Record screen-share videos to replace or augment meetings, or for announcements or status updates.

Conclusion

Provizio’s adoption of these progressive work practices is not just a response to modern trends but a well-considered strategy informed by extensive research in workplace psychology. By fostering autonomy, embracing agile methodologies, and promoting a healthy work life balance through innovative scheduling, Provizio is not only enhancing employee satisfaction and retention but is also driving the creation of higher quality products.

In the evolving landscape of work, Provizio stands as a testament to the power of aligning workplace practices with targeted, informed research, showcasing how a modern approach to work can lead to substantial benefits for both employees and the organisation.

According to the World Health Organization (WHO), there have been 1.2 million Covid-19 deaths worldwide, 44 million cases and the cost to the global economy is estimated to hit $1 trillion.

The world has rallied. United in the belief that this level of carnage is not acceptable we are taking draconian measures to stop the spread and investing billions in the science that will ultimately save millions of lives.

The automotive pandemic

According to the WHO, each year there are over 1.35 million road deaths worldwide, over 50 million people maimed and the cost to the global economy is estimated to be over $2 trillion (Link to the study here: WHO)

The numbers quoted above have not changed for more than 20 years. Unless action is taken, they will continue their current trajectory. Preventable human error is responsible for more than 90% of road accidents.

International regulatory policy for the last 20 years has focused on trying to make us better, more responsible drivers. This approach has failed consistently.

The focus by automakers on driver comfort and reactive safety measures has also, predictably failed.

Autonomous driving will ultimately solve the problem but has failed to move beyond level 2 autonomy despite $80 billion invested and will continue to do so while the roads are shared.

The vaccine

We started Provizio to solve the global road death pandemic. We believe that with the right focus, robotics and drivers can work together to reduce both road deaths and accidents to zero. 1.35 million to zero drives everything we do.

We are building augmented, guardian angel technology that will make us all better and safer drivers. We are using unparalleled ‘beyond line-of-sight’ sensor technology coupled with artificial intelligence ‘on-the-edge’ to perceive, predict and prevent accidents.

We are partnering with the top automotive OEMs and Tier 1’s to bring this capability to the masses.