TuSimple is Developing a Self-Driving Truck

Self-driving trucks are speeding towards reality thanks to the American company.


As companies like Uber, Google and Tesla take their self-driving cars out onto the open road, questions arise around the possibilities that such technology creates.

One of the challenges consists of correctly recognising lanes. Source: TuSimple

An oft-claimed advantage of using autonomous vehicles is that of the removal of ‘human error’ from driving, the cause of 94% of all accidents according to the Association for Safe International Road Travel.

TuSimple, a technology startup based out of San Diego, California, aims to dramatically reduce this number by developing an artificially-intelligent truck.

Dr. Xiaodi Hou, one of the co-founders of TuSimple, has been coding algorithim around the project since 2015. His efforts will form the core of a deep learning AI system that is at the foundation of his company’s vision.

As part of their unique approach to disruption, TuSimple’s data scientists are extending the invite to the public to get involved in the process.

Computer vision researchers are encouraged to use their own AI driving systems to analyse a dataset of 1-second videos designed to simulate real-world continuous driving conditions for a self-driving car.

The challenge consists of ‘detecting painted markings on the road that delineate each driving lane, and estimating [the] relative velocity and distance of other vehicles in view’.

The winning teams will not just be helping to advance the field of autonomous vehicles, but will also receive monetary compensation for their efforts.

Another challenge consists of recognising adjacent cars. Source: TuSimple

First place will win $1000, second place $500 and third place $250, with participants having just over a month to enter their results before the competition closes on the 21st of July.

Dr. Hou sees the challenge as not only an opportunity for learning, but also for imagination.

“Benchmark dataset competitions foster innovation and creative solutions.

“We believe that development in the field of AI is a collective process, not a zero-sum game, and any growth in knowledge or processes improves the entire field of study.”

Hopefully, by the time the challenge is completed, TuSimple will be one step closer to lowering the figure of 1.3 million lives that automobile accidents currently claim worldwide.