Engineering news
Tesla has updated its semi-autonomous Autopilot system so it uses information gathered not only from vision sensors but also from onboard radar. The aim is to make the technology more aware of its surroundings and better able to read what could be dangerous situations more clearly.
After a fatal accident involving one of the firm’s Tesla Model S sedan’s, questions had been raised about the vehicle’s semi-autonomous functionality, and how able it was to recognise objects and differentiate between dangerous and non-dangerous situations.
Version eight of the firm’s software allows the system to use the radar technology that was added to all Tesla vehicles in October 2014 as part of the Autopilot hardware suite, but was only meant to be a supplementary sensor to the primary camera and image processing system.
Tesla said: “After careful consideration, we now believe it can be used as a primary control sensor without requiring the camera to confirm visual image recognition. This is a non-trivial and counter-intuitive problem, because of how strange the world looks in radar. Photons of that wavelength travel easily through fog, dust, rain and snow, but anything metallic looks like a mirror. The radar can see people, but they appear partially translucent. Something made of wood or painted plastic, though opaque to a person, is almost as transparent as glass to radar.”
Another issue with radar is that any metal surface with a dish shape is not only reflective, but also amplifies the reflected signal to many times its actual size. So even a discarded metal can on the road, could in some instances appear to be a larger obstacle than it actually is.
“Therefore, the big problem in using radar to stop the car is avoiding false alarms. Slamming on the brakes is critical if you are about to hit something large and solid, but not if you are merely about to run over a soda can. Having lots of unnecessary braking events would at best be very annoying and at worst cause injury,” said Tesla.
Tesla said that it has faced this challenges by creating a more detailed point cloud. The latest generation of the Autopilot software can access six times as many radar objects with the same hardware with a lot more information per object.
Tesla said: “The second part consists of assembling those radar snapshots, which take place every tenth of a second, into a 3D picture of the world. It is hard to tell from a single frame whether an object is moving or stationary or to distinguish spurious reflections. By comparing several contiguous frames against vehicle velocity and expected path, the car can tell if something is real and assess the probability of collision.”
The third area to receive attention is in situations where the car is approaching an overhead highway road sign positioned on a rise in the road or a bridge where the road dips underneath, which can be read as a collision course by the system.
“This is where fleet learning comes in handy. Initially, the vehicle fleet will take no action except to note the position of road signs, bridges and other stationary objects, mapping the world according to radar. The car computer will then compare when it would have braked to the driver action and upload that to the Tesla database. If several cars drive safely past a given radar object, whether Autopilot is turned on or off, then that object is added to the geocoded whitelist,” said Tesla.