Tesla cars
Figure 6- Visual representation of camera vision. ( Future of driving , 2021)
Figure 6 will help to visualize the coverage of each of the cameras used on the Tesla vehicles.
As well as having 8 front-facing cameras, Tesla cars also have a front facing radar which can see 180m away. In its most basic form, radar simply sends out radio waves in short pulses which are reflected by objects in their path back to the radar. This means that the car can determine the distance, angle, and velocity of objects
within its range ( Radar , n.d.). Tesla use RADAR over another widely used sensor in autonomous vehicles called LiDAR. The difference in LiDAR, is that instead of sending out radio waves, it sends out light pulses in the form of a laser which bounce off a target and back to the LiDAR sensor. According to Dhiya Hani (Hani, 2021), while LiDAR produces more accurate images they are actually much less reliable due to underperforming in bad weather as well as requiring high maintenance. Radar is also much cheaper making it more worthwhile to use. Elon Musk himself said ‘If LiDAR was free, we would not put it on. ’ Furthermore, another sensor that Tesla uses are ultrasonic sensors which are fitted all around the car and each have a range of 8m. Their main purpose is to detect nearby cars and help with parking by sending out high frequency sound waves which are reflected by nearby objects and the data is collected. With all these features combined, Tesla cars can reliably understand the environment around them. While this technology provides great input, all the data produced needs to be processed. This is where the processor known as the ‘best chip in the world’ by Elon Musk becomes extremely important. The 260mm2 chip which consists of six billion transistors can carry out 26 trillion operations per second which means it is able to absorb all the information that is given to it and respond in real time to any situation that occurs on the road. A huge advantage that Tesla has over all self-driving car companies is its huge access to data. Each chip that is planted into a Tesla car will send all its data learned from on road experiences straight to HQ. With over a million Tesla cars on the road this means that huge amounts of data are getting collected every day, not only increasing the reliability and safety of the technology used in the autopilot feature, but also helping to collect important information in the transition towards full level 5 automation (highest level of automation).
Full autonomy
As part of a 5-level system to explain different levels of automation Tesla cars currently sit at level 2 where the driver still must have their hand on the wheel most of the time, while the car is able to assist in most aspects of driving. However, with heaps of data, Tesla theoretically has the capability of being in an almost fully autonomous level, though the technology does not support them being in a higher level, legally or
216
Made with FlippingBook interactive PDF creator