Ford has shown off what it's developing behind closed doors at its relatively new Innovation Centre in Palo Alto.

As part of the car company's push to become more software focused the centre was built two years ago. Since then developments have advanced in its cars including smarter Sync voice controls, autonomous vehicle development and more.

We visited the centre to see what Ford is working on now that should start to appear in its new cars soon.

Lots of these projects are aiming to have in car results by as early as the end of 2016.


Voice control smarthome from the car

Ford has teamed up with the voice controls of the Amazon Echo to allow for audio control of things outside the car.

Ford says there will be the ability to make a request in the car to activate something in the home like turning on a light for security when you've gone out. This will also work vice versa – ask how much range is available on the car while it's elsewhere, for example. This extends to all MyFord Mobile app controls like finding where the car is, turning it on, unlocking it remotely and more.

It'll be possible to ask Alexa, the Ford virtual assistant, to close your garage at home while you're away. Of course this is based on you owning an electric garage door. But it shows the start of what's possible as the car becomes connected to the smarthome.

At the moment Ford has not come up with a way to stop a car thief also getting access to your home. It says this is still a proof of concept and when this becomes a reality, later this year, those security issues will have been resolved.


Autonomous driving

Testing can be done more safely and more quickly using virtual simulation testing. This is what Ford is doing to compliment all the physical car testing from its 30 car self-driving fleet.

Ford is using a test environment that can be pushed out onto iOS devices to track behavioral responses of human drivers. This data can be logged, analysed, improved and used to update the app which is then pushed out for more testing.

Virtual testing can then be used where subjects are added to the scenario to see how autonomous cars avoid that object. Things that are focused on are the most common scenarios, so accidents, which happen rarely, don't actually need as much time and focus, says Ford.


Object recognition

The Perception Labs part of the Ford Innovation Centre works with LiDar, radar, cameras, GPU, maps and more to help cars to see. The combination of all this is called sensor fusion, something Ford hopes to perfect to create the safest self-driving car system.

A machine learning algorithm used to recognise pedestrians has been developed using just a camera. Drive DX 1 takes camera images, applies an algorithm to detect pedestrians and, in this case, pushes that back out with blue boxes around the recognised person.

But how much data and confidence can be put into a single sensor, in this case a camera? Cameras are affected by lighting so by combining it with a radar the situational awareness becomes more accurate. Testing will help to clarify this. Using many sensors is best so sensor fusion is the ultimate goal being worked on here.

The car needs a 360-degree view that recognises everything from people to trees. The ultimate goal is to illuminate all false alarms across the many sensors.

Riders for Health

Healthcare is already delivered via a fleet of vehicles in Africa by the Riders for Health charity. Over one million journeys have taken place so far with zero breakdowns. The plan was to use smart data to work out how to do this efficiently, and to build mapping data at the same time.

Plug-in in an Open XC plug-in device Ford has been tracking data from the four wheeled Ford vehicles in Africa. Ford has also developed a sensor kit that attaches to any bicycle which it's adapted for use with motorcycles.

This can be helpful for mapping data to find regular routes taken in areas with few maps.

GPS, distance, environmental, accelerometer, engine rpm, temperature and more sensors are utilised for data. The Open XC also offers air quality, traction controls and allows for characterised terrain data from the cars.

All this data will also help develop better bicycle trackers that offers the best routes for cyclists. This suggests Ford is hoping to move into all vehicle mapping so it can offer the best mapping service, perhaps as part of its new Ford Pass arm of the business.

READ: 2016 Ford Mustang GT first drive: Muscle over mileage