MIT Spinoff’s Computational Cognitive Science Technology Making Unfamiliar Conditions & Complex Interactions Redundant For Autonomous Driving

Different Levels

The thought of 5G-enabled, AI-powered self-driving cars on our roads — or autonomous vehicles (AV) to those more technically minded — is an exciting prospect for many people. And with current predictions, it is forecasted that by 2030 one in 10 vehicles will be fully automated.

However, the caveat to all this is that the technology is extremely complex. To measure this, a framework of standards for measuring AV capabilities has been formulated, with 0 (fully human-controlled) to 5 (fully autonomous) the range.

Currently, Waymo — Google’s self-driving car company — operates at Level 4 autonomy. Proof of this has come in the form of its driverless cars transporting passengers around the city of Phoenix, Arizona since 2017.

For those with a more sober and down-to-earth view about the future of autonomous transportation, partially autonomous cars (Level 2 autonomy), seem like a better and more realistic bet for what we can expect in the next ten years or so.

Yet, those special innovators and entrepreneurs working in Deep Tech, and in particular in the fields of AI and ML, want nothing to do with these level-headed predictions.

For they want the future of driving and mobility now.

ISEE, an MIT spin-off based in Boston founded in 2017 by Yibiao Zhao, Debbie Yu and Chris Baker, is one such contender accelerating the future of autonomous vehicles with its cutting-edge technology inspired by computational cognitive science that can deal with “unfamiliar situations and complex interactions on the road.”


The startup’s breakthroughs in AI and engineering has created

• Smarter autonomous driving

ISEE AI is designed to interact with other people and equipment in unexpected situations — delivering the versatility needed to automate complex operations without disruption

• Integrated engineering

ISEE brings its autonomous vehicles to life in real-world, industrial environments with robust software and hardware engineering capabilities.

These two factors bring about AI designed for flexible autonomy that can understand uncertainty, predict human patterns of behaviour while, at the same time, being able to calculate risk and efficiency.

“ISEE is reverse-engineering human intelligence to make self-driving technology that works better and safer than the kinds of things we have today.”

 — Josh Tenenbaum, Professor at MIT, Chief Scientific Advisor at ISEE

One of the interesting use cases of the ISEE’s technology is its “Yard Solution”, launched in June of this year. A yard is a place at the intersection between warehouses and logistics and is a crucial link in transportation management practices. If things go wrong in a yard, it could have catastrophic effects on the efficiency of the supply chain farther down the line.

The Yard Solution solves this by “keeping freight moving in and out of […] yard[s] quickly and efficiently while keeping other drivers and employees safe.”

The first yard truck and semi-truck employing ISEE’s autonomous driving system was rolled out in 2018. Now, three years later, the startup is expanding its fleet and deploying its state-of-the-art technology in yards throughout the United States.

On the platform and its launch, Yibiao Zhao, Co-Founder and CEO of ISEE, said:

“Unlike other autonomous vehicle solutions, ISEE AI anticipates unexpected behavior, flexibly adapts to any environment, and works alongside human driven vehicles without disruption. Our solution understands its surroundings and intuitively predicts other drivers’ behavior, enabling smooth vehicle-vehicle and human-vehicle coordination, keeping equipment and drivers safe.

Our launch marks the beginning of our mission to build towards a future where autonomous machines can thrive alongside humans and easily and safely integrate into any environment.”

For this and other ventures ISEE has in the pipeline, cash is a critical consideration, and so far the startup has raised a total of $17.7 million in funding over two rounds to charge its quest in autonomous AI for the logistics space.

We’ve looked at the technology, applications/use cases and money matters, so now let’s examine the founding team for a moment, responsible for the wonderful IP and vision.

Yibiao Zhao is ISEE’s CEO. Formerly a Postdoctoral Research Associate at MIT, he obtained a Ph.D. in Computer Vision, AI, Robotics, and ML from the University of California, Los Angeles. Here’s an interesting 2018 keynote from him at the MIT Quest for Intelligence Launch: Engineering Common Sense:

ISEE’s COO is Debbie Yu. Starting off her career as a Business Strategy Analyst at Accenture, Yu has worked in banking, as a venture growth and early-stage investor, as well as cofounding Spatial Intelligence, an environmental data solution startup specialized in the integrated use of satellite data, air quality models and ML to characterize air quality at multiple temporal and spatial scales. Yu received a BA in Economics from the Harbin Institute of Technology and an MBA from Seoul National University.

Chris Baker makes up the trio of ISEE’s co-founders and is the startup’s CSO. He gained a Ph.D. in Cognitive Science from MIT.

The spinoff’s technology is an exciting prospect for both the logistics and the autonomous vehicle industries, respectively, and with the tech developing more quickly than journalists can put pen to paper, it will probably be in no time that we see more of this kind on our roads.