STEX25 Startup:
May 11, 2021 - October 31, 2023

Common sense artificial intelligence

By: Daniel de Wolff

Well before the Suez Canal delays and the compounding effects of the pandemic, supply chain logistics were buckling under consumer demand, inefficiencies caused by unprecedented and escalating delays, and a shortage of skilled truck drivers. And while the pandemic has accelerated timetables for warehouse automation and robotics applications, and companies continue to invest heavily in their over-the-road fleets, one aspect of the logistics value chain has been consistently overlooked.

The distribution yard, where containers of goods are offloaded and stored before being trucked out to their various destinations—the shelves of our grocery stores, carmakers, farmers, and builders—is an analog anachronism in a digital world. Time and money are lost to chaos as humans search for containers, yelling into walkie-talkies, scribbling notes on clipboards, all while attempting to navigate congested yard lanes. It’s inefficient at best, not to mention dangerous for workers on location.

Early on, we identified the logistics industry as a space that needs modernization and optimization to meet the growing needs for the rapid and efficient movement of goods.

Enter ISEE, the MIT spinout improving yard efficiency and safety with a flexible autonomous solution built to bring logistics into the 21st century. “Early on, we identified the logistics industry as a space that needs modernization and optimization to meet the growing needs for the rapid and efficient movement of goods,” says Yibiao Zhao, Co-founder and CEO of "The yard still runs on old technology, but the space is critical, and can be the bottleneck of the entire value chain.”

With Zhao at the helm, ISEE has created an autonomous system that anticipates unexpected behavior, human and otherwise, so its customers can automate their existing yards without disruption while reducing costs through improved cycle times. The result is a more efficient, predictable, and safer yard that helps products get where they need to go faster and at a lower cost.

Zhao did his post-doctoral research with Professor Joshua Tenenbaum, who leads the MIT Computational Cognitive Science Group. Tenenbaum, who now serves as ISEE’s Chief Scientific Advisor, has devoted much of his career to uncovering the logic behind our everyday inductive leaps. His work combines the complementary goals of achieving a better understanding of human learning in computational terms and building computational systems that come closer to the capacities of human learners.

Meanwhile, one of Zhao’s co-founders, fellow MIT alum Chris Baker, now Chief Scientist at ISEE, pursued a line of research at the Institute that saw him developing computational models of human psychology. While the link between autonomous robots and human psychology might not be immediately apparent, Zhao points out that their research at the Institute that led to ISEE's humanistic AI is the missing piece of the autonomous driving puzzle.

“Robots today are designed to avoid obstacles,” Zhao explains. “But human drivers and pedestrians are more than obstacles, they are intelligent agents. In fact, humans are the greatest challenge when it comes to designing robots capable of operating in unconstrained environments. Understanding human decision-making processes and intention is essential.”

We basically turn our customers’ trucks into autonomous vehicles.

Reverse engineering human decision-making processes, working at the intersection of human intelligence and artificial intelligence, allows Zhao and ISEE to instill their robots with, for lack of a better term, common sense, thereby opening the door to a future where autonomous vehicles can operate in open, complex environments populated with the most unpredictable agents of all: humans.

It’s a differentiated autonomous driving system built for the real world. And despite the complexity of the technology, adapting ISEE’s tech into existing workflows is relatively simple. “Using our solution is just like hiring another driver,” says Zhao. “We basically turn our customers’ trucks into autonomous vehicles.” The result is a groundbreaking solution that integrates seamlessly into any customer’s shipping yard to streamline operations and keep freight moving in and out of the yard quickly, safely, and efficiently.

All of which has helped attract funding from big players that recognize the potential for substantial market impact. ISEE’s seed round of funding was led by The Engine, the venture fund founded by MIT to support tough tech. Shortly thereafter, Founder’s Fund joined the mix, leading ISEE’s series A. It was the Silicon Valley VC’s first push into the automated shipping and logistics industry.

Zhao sees joining the ranks of MIT Startup Exchange’s STEX25 accelerator as further proof that ISEE is on the right path. “We’re very proud of our MIT roots,” he says. “The Institute has played an essential role in shaping our technology and supporting our growth. Being named to the newest cohort of STEX25 startups is an exciting opportunity for us further connect with corporates while participating in an active community of game-changing startups from the MIT ecosystem.”

The Institute has played an essential role in shaping our technology and supporting our growth.

To date, ISEE’s customers include Maersk, the world’s largest supply chain distribution and transportation company, and Lazer Spot, the largest yard operator of distribution centers in North America. “We’re focused on deployment,” says ISEE Co-founder and COO Debbie Yu. "We're an industry-ready company with existing customers, and we have very clear goals to scale up our deployment within their networks. The future for us is more trucks, more shifts, more sites, more customers.”

It’s also apparent that updating the shipping yard is just the beginning for Zhao and ISEE. By 2050, Zhao and his colleagues expect their revolutionary platform will play an integral role in realizing automated delivery of products to customers in their homes. Consider the current iteration of ISEE the logical and likely critical step towards a future where autonomous machines can thrive alongside humans and seamlessly and safely integrate into any environment.