TuSimple Autonomous Semi navigates through Shanghai’s YangShan Deep-Water Port Logistics Park as … [+]
TuSimple announced that it successfully performed China’s first fully autonomous trucking trials without a human driver. It follows a previous announcement that it had been granted a license to conduct such tests by the Pudong New Area of Shanghai. The trials were performed by TuSimple China, a wholly-owned subsidiary of TuSimple, a U.S based corporation listed on the Nasdaq (Nasdaq: TSP). TuSimple went public in early 2021, and as opposed to others in the autonomous driving space who went the SPAC (Special Purpose Acquisition Company) route, it did this in an old-fashioned way via an IPO (Initial Public Offering). It is currently valued at ~$$500M and has ~$1B in cash. Revenues are modest (~$10M/year).
The company has demonstrated completely autonomous trial runs in the United States with a fleet of over 50 trucks that have logged upward of 3 million test miles hauling loads from depots in Tucson, Arizona, and Dallas with human safety drivers. In December 2021, TuSimple, completed the world’s first fully autonomous semi-truck run on open public roads without a human in the vehicle and human intervention. It occurred between a large railyard in Tucson to Phoenix, Arizona, a distance of 80 miles on surface streets and highways. Recently, it has begun Level 4 autonomous test runs on Japan’s most critical freight corridor that connects the major cities of Tokyo, Nagoya, and Osaka.
The tests in China were conducted over a 62-kilometer (39 miles) route linking the Yangshan Deep-water Port Logistics Park and Donghai Bridge that connects the port to Shanghai. The route includes urban and highway traffic and requires recognizing and negotiating on and off-ramps, lane changes and traffic signals. Fog and high crosswinds are also complicating factors. TuSimple was one of four companies (SAIC Motors, a prominent automotive OEM in China, is one of the others) to be granted a license for this testing but appears to be the first to conduct this trial successfully.
Yangshan Port and Donghai Bridge
The Yangshan Port completed construction in 2016 and is the world’s largest and busiest deepwater port, designed to service large and heavy cargo ships. It is part of the Shanghai Port system and handles upward of 40 million TEUs (20-foot containers) annually. For reference, large ships carry ~20,000 such containers; typical trucks carry between 1-2 TEUs depending on their weight. At total capacity, this requires between 60-120 K truck journeys per day across the system. The port is connected to the mainland (Shanghai) via the Donghai Bridge, a 6-lane highway (3 lanes in each direction), and is one of the longest trans-oceanic bridges in the world.
SHANGHAI, CHINA – AUGUST 14: Aerial view of Donghai Bridge which connects Shanghai’s Pudong New Area … [+]
Impact of Fully Autonomous Trucking in China
TuSimple’s CEO Cheng Lu: “Being the first to conduct a Driver Out run in China is a significant milestone. Following our successful Driver Out tests in the United States in 2021, this accomplishment marks another pivotal breakthrough for TuSimple and further underscores our leadership in the autonomous driving industry. The benefits include safety, energy and throughput efficiency in critical logistics links at some of the busiest ports in the world. Successful autonomy also solves significant driver shortage issues in many parts of the globe”.
The freight market in China is ~$700B/year and is addressed by ~ 6M Class 8 heavy-duty trucks (the United States market is similar and handled by 2.5M trucks). The potential for addressing safety and economic efficiency is significant for autonomous trucking companies like TuSimple. However, the competition is severe (>20 companies are in the game), and regulatory, public acceptance, labor relations and technical challenges are complex and require significant investments and patience.
TuSimple Approach for Perception
Apart from AI (Artificial Intelligence) algorithms based on millions of miles of driving and simulated data, a key technical challenge is the perception stack which is critical for understanding the driving environment around the vehicle. Along with cameras and radar, TuSimple uses LiDAR (Light Detection and Ranging) as a crucial element in controlling the drive stack that manipulates the control actions of the truck (lane changes, acceleration, braking). Heavy trucks moving at highway speeds require long-range perception (1 km). Cameras and radar provide this. Current LiDAR units enable 300 m of perception range, and TuSimple uses a combination of LiDARs for short and long-range perception. FMCW (Frequency Modulated Continuous Wave) LiDAR is combined with classical ToF (Time of Flight LiDARs) to sense the range and the radial velocity of surrounding vehicles. Innovusion, Aeva and Aeye are the key LiDAR suppliers. Vehicle styles in the United States and China are different, and the perception stack (hardware and software) is customized for local conditions.
Relative to the peak in 2015-2017, the autonomous vehicle industry is undergoing challenges and growing pains. TuSimple is not an exception. Since the heady days of its IPO in early 2021, it has gone through management upheavals that have resulted in CEO and founder firings and returns. U.S. regulatory scrutiny has been significant, which led to discussions about the China operation being spun off (although CEO Cheng Lu indicated this is no longer the case). Nasdaq threatened to de-list the company due to financial reporting gaps. Rounds of layoffs have reduced its headcount from a peak of ~1000 to ~250 today. Market capitalization has dropped from ~$10B in 2021 to ~$500M today. An agreement with Navistar for autonomous truck development in the United States has been canceled (although Navistar is still an investor). The company recently announced that it is exploring strategic options for its North American business, one of which could be a sale. This development is likely driven by the need to focus its cash and human resources on capitalizing on its opportunities in Asia Pacific (China and Japan) and reduce the U.S. political/regulatory headwinds regarding intellectual property and information sharing with China.
The market opportunities for autonomous trucking are compelling because of the potential to improve efficiency, safety and profitability and address driver shortage issues. TuSimple is a pioneer in this space with an impressive track record of autonomy trials in the United States, China and Japan. Hopefully, they can emerge as a winner in this increasingly complex and competitive space.
