What you will learn:
- Details of the Toyota/Pony.ai Sienna-Automo Maas robotaxi.
- The basis of the system: the DRIVE Orin from NVIDIA.
- Test results and road tests to come.
Robotaxis, self-driving taxi services operated by ride-sharing companies, are fast becoming a major mobility solution. Robotaxi operations require SAE Level 4 autonomous vehicles capable of operating without a backup driver, but are often monitored by teleoperators ready to take control of the vehicle at a moment’s notice. Essentially, they are limited by specific operating conditions, such as geography or weather.
According to SAE guidelines, a Level 4 car should be able to drive safely “even if a human driver does not respond appropriately to a request for action.” A level 4 car will slow down, stop, or park in a safe place if the driver does not take control on demand, which can happen when driving off-road or on unmapped roads.
By comparison, most of today’s Advanced Driver Assistance Systems (ADAS) fall under Level 2, including Tesla’s Autopilot, Cadillac’s Super Cruise, and Volvo’s Pilot Assist. While Tier 1 vehicles control either speed or direction, Tier 2 vehicles can control both simultaneously.
Recently, the Chinese autonomous vehicle startup pony.ai announced that its next-generation robotaxi is built on the Nvidia DRIVE Orin platform. The computing unit will be manufactured on a standardized production line. The first model equipped with this system, Toyota’s Sienna-Automo Maas, is a seven-seater hybrid electric vehicle that will begin road tests in China this year and will be deployed in Pony.ai’s public robotaxi operations during the first semester 2023.
Toyota and Pony.ai began jointly developing specially-designed Sienna vehicles for robotaxi services in 2019. The customized vehicles feature a dual-redundant system and better control performance for Level 4 autonomous driving capabilities. “We have accumulated over 10 million miles of self-driving mileage globally on open roads and throughout 2022 we will continue to invest in our core R&D and global robotaxi operations,” said James Peng, co -founder and CEO of Pony.ai.
In the S-AM model, Pony.ai unveiled a new design concept for its autonomous driving system, making the sensor suite more integrated and aesthetically pleasing while being more suitable for mass production. Among other design features, it includes roof signaling units that have a horizontal lighting unit at the front and three vertically placed light strips at the rear. By using different colors and lighting combinations, rooftop signaling units can provide external communication and interaction, demonstrating the status and intentions of the robotaxi.
Toyota’s sensor solution, containing a total of 23 sensors, includes four solid-state LiDARs on the roof covering a 360-degree field of view; the central mechanical LiDAR has been replaced by solid-state LiDARs. Three short-range LiDARs on the vehicle body cover the blind spots of the roof LiDARs. Additionally, four millimeter wave radars sit at the corners of the roof, and there is a long-range forward-looking millimeter wave radar. Additionally, 11 cameras are deployed around the roof and body of the vehicle (in a combination of wide-angle, super-wide-angle, mid-range, long-range and traffic-light detection cameras).
The in-house developed traffic light camera has a 1.5 times higher resolution rate than the previous generation device. Coupled with its in-house sensor fusion technology, Pony.ai said it would significantly reduce the costs associated with the solution by using mass-produced automotive-grade sensors.
Pony.ai has done road tests with the DRIVE Orin. Compared to the previous computing platform, the next generation is expected to have more than 30% increase in computing power, at least 30% less weight and more than 30% reduction in cost.
Pony.ai has government permission to conduct driverless testing and charge fees for public robotaxi services in Beijing, as well as permits to test self-driving vehicles in many major Chinese cities. Last year, Pony.ai became the first company to conduct self-driving trucking tests on an open-air highway in China, and is the only company to hold self-driving trucking permits in Beijing and Guangzhou.
NVIDIA’s DRIVE Orin achieves 254 TOPS (trillion operations per second) performance and includes support for the CUDA toolchain and NVIDIA deep-learning accelerator (NVDLA). It is designed to handle the large number of applications and deep neural networks running concurrently in self-driving trucks, while adhering to systematic security standards such as ISO 26262 ASIL-D.
Pony.ai’s relationship with NVIDIA dates back to 2017, when the company first adopted the DRIVE platform. Pony.ai is one of the first in the autonomous vehicle industry to create a product portfolio featuring multiple configurations with single or multiple DRIVE Orin processors and automotive-grade NVIDIA Ampere architecture GPUs. This is said to allow for scalable deployment on self-driving trucks and robotaxis.