With the rapid advancement of artificial intelligence, a wide range of AI chips and software algorithms have seen significant progress, and the concept of self-driving cars is gaining more traction. At the recent Mobile World Congress in Barcelona, Huawei surprised attendees by not unveiling a new smartphone. Instead, it showcased its Mate10 Pro as a key focus, demonstrating how it can control a Porsche.
The Huawei Mate10 Pro, equipped with its own NPU (Neural Processing Unit), is capable of intelligent analysis of road conditions and can guide autonomous driving systems on how to respond. During a live demonstration, three obstacle boards—representing a bicycle, a dog, and a football—were placed on the road. The Mate10 Pro successfully identified these obstacles using its NPU-powered image recognition, ensuring safe navigation for the vehicle.
This was just one example of how AI is being applied in the field of autonomous driving, and it’s possible that Huawei may soon enter the self-driving car market. In fact, global automakers have made considerable progress in autopilot technology over the years. Tesla introduced Autopilot 2.0 in 2016, claiming it had reached the fifth level of full automation.
Car manufacturers are increasingly investing in autonomous driving. According to reports, Tesla has been collaborating with NVIDIA to develop the next-generation computing platform for self-driving cars. Recently, Elon Musk mentioned that Tesla is testing a major update to its Autopilot software, which is now in the final stages of development.
The system uses an NVIDIA DrivePX2 chip, along with eight cameras, twelve ultrasonic sensors, and an enhanced front radar, providing a 360-degree view of the surroundings. Tesla also employs some of the most advanced AI neural networks for consumer products, and Autopilot 2.0 is expected to be featured in new Model 3 vehicles.
In California, the state recently announced that automakers and tech companies can test their self-driving vehicles on public roads without a human driver. This marks a significant step forward, allowing companies like General Motors and Waymo to continue their testing efforts.
Waymo, a subsidiary of Google, partners with Fiat Chrysler, Lyft, and Avis, conducting tests of autonomous vehicles across various automotive companies. Waymo has even filed a patent for a motion sickness prevention system, which will automatically adjust driving patterns to avoid congested areas, reduce stops, and improve passenger comfort.
Ford has also entered the autonomous driving arena after acquiring ArgoAI, with plans to launch a self-driving model by 2021. Recently, Ford acquired a patent for driverless police car technology, which allows the vehicle to autonomously detect and track illegal vehicles, record violations, and issue penalties.
Ford is also set to test two of its latest self-driving cars in Miami, with plans to introduce a driverless delivery service. Urban delivery with autonomous vehicles presents unique challenges, but in the future, we may see more smart cars on the streets, handling both passenger transport and cargo delivery.
Despite these advancements, power consumption remains a major challenge. Electric vehicles are becoming more popular due to their zero emissions and quiet operation, but smart cars require power for everything from motors to sensors, communication systems, and entertainment. Future smart cars will need to address two key issues: increasing battery capacity and reducing energy consumption.
Advanced AI and deep learning are central to smart car technology, but running AI chips consumes a lot of power. Currently, NVIDIA chips are widely used, while Intel and Qualcomm are developing low-power alternatives. Tesla is also working on its own next-gen autopilot chips. Although these developments show promise, there's still a long way to go, as current systems use up to 2,500 watts per second.
According to a University of Michigan study, energy loss in autonomous systems accounts for 41% of total usage. Adding rooftop sensors, like those used in Waymo's self-driving cars, increases air resistance and adds weight, further impacting energy efficiency. However, fully autonomous vehicles may not be far off, as Waymo has managed to cut sensor costs by 90%.
With rising traffic accidents and carbon emissions, new energy vehicles and autonomous driving are likely to become more widespread. While affordable and energy-efficient autonomous transportation is still a work in progress, limited autonomous driving capabilities are already feasible today.
10L Agriculture Drone,Spraying Drone,Uav Folding Drone,Uav Drone Crop Sprayer
Xuzhou Jitian Intelligent Equipment Co. Ltd , https://www.jitianintelligent.com