Tesla autopilot update

The autopilot software built in the Tesla electric cars has offered us a small glimpse of how the future may look. Being inside in what seems to be a self- driving vehicle is an experience which humankind has waited for a very long time and it is only a few years before many of us can enjoy. Meanwhile, Tesla has announced a suite of automotive upgrades to its line of electric cars, and this is good news considering many consumers are pushing for cars which can reduce our over reliance on petroleum energy. The company’s CEO, Elon Musk has praised the announced upgrades as improvements which he wished were made earlier.

The autopilot feature works by building a picture of the world around it with the help of a forward-facing camera mounted on a vehicle’s rear view, special radar placed at the front bumper and twelve ultrasonic sensors which placed around the car. Until recently, Tesla electric cars have been using camera images as the primary information for judging on how a vehicle’s real-life environment may look. According to Tesla’s chief executive, the burden of forming new images will now be shifted to the radar system. The radar will unlock information on an object by unlocking as many six times the pulse reflected from the object.

The radar will ping ten times, and from this action, it will be possible to develop 3D images from received reflections the same time the car will be moving. In many instances, radar is considered not that sharp for image processing. However, Tesla has promised to make signal processing to be quality and make the involved procedure work. Repetitive receiving of object signal is seen as an improvement as this increases a vehicle’s confidence that any reflected pulse can indeed be regarded as a real object. One limitation however exists- false positives may mean a vehicle is making unexplained brakes which can turn out disastrous.

Perhaps the most important aspect that turns out is that cars cannot in any way learn on their own. Tesla’s boss, Elon Musk, further added that the company is adding 1.5 million miles per day to its autopilot. The meaning is that Tesla’s electric car fleet will be learning simultaneously at once. The radar system will be geotagged, and other users will share information collected by a given electric car. Information sharing among Tesla’s electric cars is expected to enhance safety as all the data collected will be optimized to mark the unsafe points on particular roads.

Radar has being praised as being accurate in identifying objects on the road. The explanation is arrived at by comparing the working mechanism installed in the Tesla’s electric car model that was involved in a crash in Florida with the new radar system. In the crash, the camera system mounted on the car was unable to identify a white track against the bright sky accurately. The radar will be an improvement given that object identification is more accurate as compared to relying on camera images.

Amanda Kremer