Rapid Application Development
CAVLab is the cornerstone of our technology - a sophisticated, in-house object based graphical programming language developed specifically for the rapid development of self-driving vehicles.
Building applications in CAVLab is achieved by simply connecting components together graphically.
Powerful editing tools allow us to create user interfaces in record time and with the Ruby programming language built in as well as the option of linking to externally written code, our developers have ultimate flexibility.
One of the key features of this amazing development tool is that it works in real-time allowing the developer to continue to code as the software runs - no need to constantly compile and test.
We can actually be changing the code as a vehicle is driving and see the effects as soon as we make the changes. We can also catch errors as they happen which results in far fewer bugs and creates robust, safer code.
CAVLab has built in modules for connecting to a vast range of external devices. This means that we don't have to re-invent the wheel every time we start a project.
We have components for interfacing to cameras, sensors, data acquisition devices, networks and many more.
CAVLab allows designers, engineers and programmers to work together on the same software project regardless of their programming experience.
A designer can be building a user interface while an engineer works on the core system and their contributions can easily be brought together or exchanged as an application is being built.
Powerful, Flexible Connectivity
CAVdaq is a multi-purpose, standalone data acquisition board that has been designed to work seamlessly with CAVLab.
Using CAVdaq we can connect to a huge range of sensors, GPS systems, 4/5G radio telemetry and CANbus devices. It has been designed to work on its own or several boards can be connected together to cater for the massive processing power required for Autonomous vehicles.
Wireless communication between vehicles and HQ is a key part of any fleet management system. To provide this link we added a 4/5G modem to a CAVdaq to create CAVcon.
The system is two way so as well as receiving telemetry we can also send control data to a specific vehicle, massages to the passengers or we can re-route a vehicle.
Accurate reliable GPS is essential for accurate vehicle localisation which is why we developed our own hardware.
By adding a GPS receiver to the CAVdaq we created CAVgps - a low cost RTK GPS system providing super accurate 1 cm positioning. This can be NTRIP (using 4G/5G for the corrections) or Radio based up to 10km in radius.
CAVdaq forms the base of two of our other technologies, CAVcon and CAVgps
Autonomous Control System
Our Autonomous Control System, CAVacs, uses mapped localisation data to control the driving of a vehicle over one of a number of predefined routes.
We use proprietary algorithms to take control of a self-driving vehicle and manage the vehicle's steering, braking and acceleration.
One of the key things about the CAVacs is that it can be quickly dropped into any vehicle, be that a car, bus, digger, pod or tractor.
LOCALISATION & MAPPING
CAVacs records fixed objects in real-time whilst localising on a known route. All routes are mapped using cm accurate RTK GPS.
Any fixed objects are added to our high definition map. We can also define the road width and add passing places, junctions and pedestrian crossings.
HUMAN MACHINE INTERFACES
Using our CAVlab rapid development system we are able to quickly tailor any HMI experience to a customer's needs. We can provide multiple touch screens displaying real-time data inside vehicle, a building or across multiple sites.
These can be used to create rolling advertising boards with sound and video as well as displaying information and controls inside the vehicle.
The Cavonix video processing system uses Ethernet HD cameras to detect and classify objects such as other vehicles and pedestrians.
Detected objects can be tracked and from the tracking data we can determine their direction of movement. This informs decision making at junctions and collision avoidance.
If objects are detected in the path of a vehicle CAVacs can make decisions about how to avoid a collision. It can either bring the vehicle to a halt or it can calculate a new route around the object. This is done whilst obeying strict rules to ensure vehicles make safe decisions.
CAVacs also uses geofencing to ensure that if a vehicle deviates off path it is spotted instantly and the necessary corrections are made.
Our CAVtrak Fleet Management System (FMS) monitors location and status for a pool of vehicles. Communication is two-way allowing you to modify vehicle routing and operating parameters in real-time.
The software was developed using CAVlab and so it can be quickly adapted to different sites and data requirements.
Autonomous vehicles need to be able to react to the world around them so we have developed a range of sensor input modules and accompanying software to make decisions on whether vehicles should adjust their course, slow down, accelerate or stop based on what is happening nearby.
Our CAVsense Sensor Fusion system uses a range of hardware from bespoke high speed microprocessor systems to powerful embedded PCs to best process the huge amount of real-time data hundreds of times a second.
Since no single sensor is sufficient to provide a complete picture we use a combination of real-time HD video with Radar, and LIDAR to detect any objects around the vehicle
Real-time RADAR data detects hundreds of objects in and calculates their relative velocities to foresee potential collisions way in advance.
We use a new type of Flash Radar unit that is very robust in many difficult conditions such as heavy rain, fog and snow. This gives a 360 degree view around the vehicle which can be used for object collision detection.
High resolution cameras complete the picture. We use these to detect nearby vehicles and predict their movements.