SE Valley Times

Fantasy writer Joanna Maciejewska recently shared her thoughts on social media platform X, stating, “I want AI to do my laundry and dishes so that I can do art and writing, not for AI to do my art and writing so that I can do my laundry and dishes.” Her post resonated widely, amassing over 3 million views and 100,000 likes.

Maciejewska's comment reflects a common sentiment about artificial intelligence (AI) as it becomes more prevalent in daily life. While tools like Midjourney are used to create artwork quickly, there is concern among educators about students using platforms like ChatGPT for assignments. The ideal scenario would be for machines to handle undesirable tasks, allowing people more time for creative pursuits.

Siddharth Srivastava, an associate professor at Arizona State University’s School of Computing and Augmented Intelligence, is working towards this vision. He leads the Autonomous Agents and Intelligent Robots Lab (AAIR Lab), where he and his team research how robots can perform complex tasks independently.

“We must find ways to get robots to do those jobs that are dirty, dull or dangerous,” Srivastava explains. “We want them to do the kinds of jobs people would prefer to avoid.”

Srivastava's work is supported by multiple National Science Foundation grants aimed at developing AI solutions in robotics. His team is creating a toolkit combining algorithms with language data and real-world models. These innovations will enable robots to devise new plans for solving complex tasks without user input.

Currently, many robotic systems require expert-level coding for specific tasks. This makes development costly and limits robot deployment in various settings. Srivastava aims to simplify this process by enabling robots like Alfred—a Fetch mobile manipulator robot—to learn autonomously.

Alfred has been learning how to clear dishes after meals by observing demonstrations of picking up cups and bowls. Using Srivastava’s algorithms, Alfred develops its own understanding of necessary actions without needing hand-coded instructions from experts.

“The task of setting a table with hand-coded primitives has been done before,” Srivastava says. “The main technical innovation here is that the robot learns on its own how to generalize.”

This approach allows robots like Alfred to operate anywhere without preprogrammed spatial information while reducing costs associated with manual programming.

Potential applications include medical settings where robots could clean hospital rooms by recognizing different configurations autonomously or disaster-recovery support systems adaptable under uncertain conditions.

Before embarking on this project, Srivastava was already exploring similar technologies envisioned by individuals like Maciejewska—such as developing a robot capable of doing laundry automatically.

“My family is eager for me to finish this project and get back to the laundry issue,” he jokes.