Robots are using game theory to keep humans safe

Picture an auto factory humming with energy. A robot rapidly assembles car doors while a human checks each one for quality. Their partnership looks seamless – with machines handling speed and strength, and people applying judgment and dexterity.

These scenes symbolize the growing integration of humans and robots across industries. But as the collaboration deepens, so do the risks.

Human error, unpredictability, and miscommunication can create moments robots are not prepared to handle – moments can have serious consequences.

Teaching robots safe choices

Researchers are now tackling this challenge. At CU Boulder, Professor Morteza Lahijanian and his team are building processes that help robots make safe yet effective decisions around people.

In a new study, Lahijanian and graduate students Karan Muvvala and Qi Heng Ho have introduced algorithms that allow robots to manage risk while completing tasks.

“How do we go from very structured environments where there is no human, where the robots are doing everything by themselves, to unstructured environments where there are a lot of uncertainties and other agents?” Lahijanian asked.

Robots predict human actions

Robots, like humans, rely on mental models to make decisions. When working with a person, a robot predicts possible actions and adjusts accordingly.

“If you’re a robot, you have to be able to interact with others,” said Lahijanian. “You have to put yourself out there and take a risk and see what happens. But how do you make that decision, and how much risk do you want to tolerate?”

The team drew inspiration from game theory, a mathematical framework originally from economics. In robotics, game theory treats each robot as a player in a game. Winning means completing a task, but with humans involved, the game gets unpredictable.

Algorithms minimize future regret

Instead of ensuring robots always win, the researchers proposed “admissible strategies.” With this approach, a robot aims to finish its job while minimizing harm. Safety remains the top priority.

“In choosing a strategy, you don’t want the robot to seem very adversarial,” Lahijanian said. “In order to give that softness to the robot, we look at the notion of regret.

“Is the robot going to regret its action in the future? And in optimizing for the best action at the moment, you try to take an action that you won’t regret.”

Robots adapting to humans

Back at the auto factory, imagine the human making repeated mistakes. Using the new algorithms, the robot might fix those errors safely. If that fails, it could relocate the task to a safer space, protecting both the product and the worker.

Much like a chess master predicting several moves ahead, robots can anticipate human choices. Perfect prediction is impossible, but proactive strategies prioritize safety.

“If you want to have collaboration between a human and a robot, the robot has to adjust itself to the human. We don’t want humans to adjust themselves to the robot,” Lahijanian said.

This flexibility allows robots to work with novices and experts alike. Regardless of the human partner’s skill level, the robot must respond safely and intelligently.

Beyond the factory floor

The auto factory is only one example. Hospitals could also benefit. Imagine a nurse and a robot sharing patient care tasks.

The robot might deliver medications or carry equipment, leaving the nurse free to focus on judgment-driven decisions. If errors occur, the robot could adjust without creating new risks.

Construction is another field where collaboration could shine. Robots could take on heavy lifting while humans manage fine-detail tasks like alignment or inspection. Agriculture also stands to gain.

Machines could harvest crops at scale while farmers concentrate on resource management and sustainable practices.

Robots ease labor shortages

When robots work safely with humans, they can provide clear benefits. Industries facing labor shortages, such as elder care, could see relief. Physically demanding jobs might also become safer for human workers.

Lahijanian emphasized that robots are not meant to replace human talent but to expand it.

“Human-robot collaboration is about combining complementary strengths: humans contribute intelligence, judgment, and flexibility, while robots offer precision, strength, and reliability,” he said. “Together, they can achieve more than either could alone, safely and efficiently.”

The research was presented at the International Joint Conference on Artificial Intelligence in August 2025.

—–

Like what you read? Subscribe to our newsletter for engaging articles, exclusive content, and the latest updates. 

Check us out on EarthSnap, a free app brought to you by Eric Ralls and Earth.com.

—–