Tesla has a long and sordid history of Fully autonomous driving capabilities, to the point that CEO Elon Musk’s announcements and assertions about how revolutionary next update will often be dismissed by Tesla fans and detractors. The latest announcement seems to be something a little different from the “full battery life in six months” that we’re used to seeing.
According to a tweet from Elon Musk, Tesla is currently working on “the final piece of the Tesla FSD AI puzzle”, which is vehicle control. Controlling the vehicle via machine learning or AI appears to indicate that Tesla is training an AI or ML neural network to control its vehicle instead of relying on sets of pre-coded instructions for steering, braking and vehicle throttle inputs.
This move to AI and ML on predefined code is expected to bring the company closer to reaching Level 5 of autonomous driving and achieving its “end-to-end AI” fully autonomous driving goal.
Vehicle control is the final piece of the Tesla FSD AI puzzle. This will remove over 300,000 lines of C++ control code by approximately 2 orders of magnitude.
This is training as I write these lines. Our progress currently concerns training under IT constraints, and not under engineering constraints. — Elon Musk, CEO of Tesla
Over the years, Tesla has repeatedly changed its fully autonomous driving strategy. One of the most significant changes in recent memory has been the removal of radar equipment moving to a camera-only system – a move that has been widely criticized as a cost-cutting measure.
It remains to be seen what an AI or machine learning-based FSD package would look like, but it could open up the possibility for FSD software to handle much more complex interactions more easily. Even after Tesla resolved the technical issues, it still faces hurdles in the form of regulatory oversight and public perception – neither of which have been positively influenced by recent incidents and accidents, such as Bay Bridge pileupinvolving Tesla’s FSD package.
My interest in technology started in high school, rooting and flashing my Motorola Defy, but I really fell down the rabbit hole when I realized I could overclock the i7 930 in my pre-built Gigabyte PC. This DIY addiction eventually led me to study product design at university. I believe technology should improve the lives of those who use it, regardless of the field. I enjoy reading and writing about laptops, smartphones, software and technology trends.
