Self-Driving Truck Developer TuSimple Completes 550 Miles of Freight Deliveries on Public Roads Without a Driver or Remote Backup

author: Eric Walz   

California autonomous truck developer TuSimple is making steady progress in its goal to deploy a fleet of long haul trucks that can operate entirely without human intervention.

On Wednesday, the company announced that Union Pacific Railroad, which is the largest Class I railroad in the U.S., will be the first customer to move freight on TuSimple's fully-automated trucking route between the Tucson and Phoenix, Arizona metro areas. 

The partnership with Union Pacific makes sense, since TuSimple compares its driverless truck routes to a railroad, with trucks following set routes like railway links between two fixed points.

"Partnering with TuSimple allows us to extend our operations beyond our rail hubs and serve our customers faster and more efficiently," said Kenny Rocker, Executive Vice President – Marketing & Sales, Union Pacific. "This groundbreaking autonomous driving technology and our partnership provide us a significant opportunity to scale the technology in our network, proactively reducing global supply chain congestion."

Union Pacific is leveraging Loup Logistics (Loup), a wholly-owned subsidiary, to coordinate the freight shipment and support seamless movement between rail and the critical first and last mile. 

Beginning in the spring, TuSimple plans to undertake what its calls are "Driver Out" freight deliveries for Union Pacific. The Driver Out deliveries mean that a driver won't be behind the wheel, nor will TuSimple rely on a remote teletoperator as a backup to take over.

"Our repeatable and scalable 'Driver Out' operations marks a significant inflection point in our company's history. We are the world's first to complete all of the features of AV trucking technology," said Cheng Lu, President, and CEO, TuSimple. "We are proud of our on-time delivery of this historic milestone and are excited to shift our full focus to commercializing our ground-breaking technology on an accelerated timeline."

TuSimple, which was founded in 2015 and is headquartered in San Diego, California. The company is working to transform the $800-billion U.S. trucking industry. According to McKinsey, full autonomous long-haul trucks could reduce operating costs by about 45% and can be utilized 24 hours a day without the long breaks required for drivers.

Last month, TuSimple announced it completed the world's first "Driver Out" semi-truck run on public roads in one of its self-driving Class-8 trucks. The company says the 80-mile trip was the first-ever successful fully-autonomous run by a class 8 vehicle, or semi, on open public roads with no human intervention. The journey began at a railyard in Tucson, Arizona and ended at a distribution center in Phoenix. 

Since then, the company said it completed another six successful fully-autonomous runs. The seven total runs covered over 550 cumulative miles on open public roads without a human in the vehicle, without any teleoperation, and without any traffic intervention, according to TuSimple. 

The autonomous freight delivery runs were conducted in various roadway conditions, including dense early-evening traffic and "back-to-back" runs on the same night. By operating at night, autonomous truck fleets can be more efficient, increasing truck utilization from an average of 50% or 12 hours per day, to an average of 80%.

A core part of TuSimple technology is its long-range, camera-based perception system developed specifically for Class-8 trucks, which can identify objects up to 1,000 meters ahead. 

Class-8 trucks take much longer to stop than a passenger car, so it is important to be able to perceive objects much further ahead in order for the autonomous driving software to react in time.

Like developers of self-driving passenger vehicles, TuSimple's long-range perception system uses a combination of LiDAR, radar and HD cameras. It allows the autonomous system to successfully navigate in nearly any driving condition, while keeping a truck centered in a lane with an accuracy of 5 centimeters, according to TuSimple.

Last month, TuSimple announced it expanded its partnership with chipmaker NVIDIA to design and develop an advanced "autonomous domain controller" (ADC) specifically for TuSimple's Level 4 autonomous trucking applications.

TuSimple's ADC incorporates the NVIDIA DRIVE Orin system-on-a-chip (SoC), which is specifically designed for AI-based autonomous driving applications. Nvidia's Orin is one of the most powerful SoCs. It delivers the computing power required for advanced AI-powered computer vision processing.

The Nvidia-powered ADC is an integral part of TuSimple's highway autonomous driving system for Class-8 trucks, supporting perception, planning, and actuation functions. It will be integrated into the company's future autonomous truck production programs. 

Going forward, TuSimple plans to continue its 'Driver Out' program and progressively expand its scope to incorporate daytime runs and new routes while making regular improvements to its proprietary AV technology. 

TuSimple plans to achieve commercial viability by 2023 with continuous 'Driver Out' paid freight operations in a significant shipping area such as the "Texas Triangle", which is the nickname for the region of Texas with the state's five largest cities and the majority of the state's population. 

The wide open roads of the Texas and Southwest U.S. are serving as an ideal proving ground for TuSimple to test its Class-8 trucks and perfect the autonomous driving technology as the company prepares for commercial deployment.

Eric Walz

Originally hailing from New Jersey, Eric is a automotive & technology reporter covering the high-tech industry here in Silicon Valley. He has over 15 years of automotive experience and a bachelors degree in computer science. These skills, combined with technical writing and news reporting, allows him to fully understand and identify new and innovative technologies in the auto industry and beyond. He has worked at Uber on self-driving cars and as a technical writer, helping people to understand and work with technology.