Tesla’s Full Self-Driving (FSD) Beta is an advanced driver-assistance system (ADAS) that aims to pave the way for driverless vehicles. The ultimate goal of Tesla is to develop a licensable product that can be sold to other car manufacturers, similar to how it commercialized the Supercharger network.
One notable capability of FSD Beta is its ability to react to emergency vehicles. In a demonstration, a Tesla Model Y equipped with FSD Beta v11.4.4 showcased its predictability and awareness of an approaching ambulance. While waiting at a traffic light, the Model Y remained stationary even when the light turned green, allowing the emergency vehicle to pass. In contrast, a human driver in a Toyota proceeded to move even with the ambulance approaching.
Although FSD Beta displays impressive behavior in this scenario, it is important to note that the system has been known to make erroneous decisions. It still requires further development and fine-tuning to fully comprehend various traffic situations better than a human driver.
Furthermore, Tesla has been refining FSD Beta to run specifically on vehicles equipped with Hardware 4. Despite being slightly behind vehicles with Hardware 3, this adjustment instills confidence in the capabilities of the ADAS.
Tesla plans to transition FSD to a neural net-based software with the release of FSD V12. This iteration will utilize customer-submitted videos to train the system. Instead of manually coding each scenario, the system learns from observing correct behaviors exhibited by multiple individuals in specific situations.
It is worth noting that Tesla’s ADAS solution relies solely on cameras and does not employ other sensors such as ultrasonic, radar, or LiDAR technology. In contrast, competitors like Cruise, Waymo, and Polestar utilize multiple hardware solutions to achieve self-driving capabilities. Polestar, for instance, will incorporate Mobileye’s suite into its vehicles from 2025, offering a more comprehensive approach beyond just cameras.
While Tesla’s FSD Beta is a promising product with potential market dominance, challenges lie ahead. Competitors like Ford and General Motors have deployed their own ADAS systems like BlueCruise and Super Cruise, allowing hand-off driving on pre-approved roads. Mercedes-Benz has also received approval for its SAE J3016 Level 3 Drive Pilot. In comparison, FSD Beta currently operates at Level 2, emphasizing the need for users to exercise caution and not rely excessively on the technology.
In conclusion, Tesla’s FSD Beta showcases advancements in ADAS technology and its ability to react to emergency vehicles. However, further development is necessary to address errors and enhance understanding of diverse traffic scenarios. Tesla’s future plans involve transitioning to neural net-based software and relying on customer-submitted videos for training. While competitors adopt multiple hardware solutions, Tesla remains committed to its vision-based approach. Challenges from other automakers’ ADAS systems and different levels of autonomy pose future obstacles for Tesla’s FSD Beta, which currently operates at Level 2.