Mon. Dec 23rd, 2024
Robotics Q&a: Cmu's Matthew Johnson Roberson

Johnson-Roberson is one of those dual threats who provides insight from two different and important perspectives. In addition to his long academic career, including most recently as a professor at the University of Michigan’s School of Engineering, he also has a solid resume in startups.

Johnson-Roberson is also the co-founder and CTO of robotic last-mile delivery startup Refraction AI.

What role will generative AI play in the future of robotics?

Generative AI greatly enhances the capabilities of robots with the ability to generate new data and solutions. This may allow students to be more generalizable across a wide range of tasks, better able to adapt to new environments, and improve their ability to learn and evolve autonomously.

What do you think about the humanoid form factor?

Humanoid form factors are extremely complex engineering and design challenges. The desire to mimic human movements and interactions creates high hurdles for actuators and control systems. It also presents unique challenges in terms of balance and coordination. Despite these challenges, humanoid shapes have the potential to reflect natural human interfaces and interactions and to be extremely versatile and intuitive to use in a variety of social and practical situations. I am. But other platforms will probably succeed before these.

What will be the next major category of robots after manufacturing and warehousing?

Beyond manufacturing and warehousing, the agricultural sector presents a huge opportunity for robotics to address challenges of labor shortages, efficiency, and sustainability. Transportation and last-mile delivery is another area where robotics can increase efficiency, reduce costs, and improve service levels. We expect the adoption of robotic solutions to accelerate in these areas as the technology matures and regulatory frameworks evolve to support broader deployment.

How far have true general-purpose robots evolved?

The emergence of truly versatile robots capable of performing a wide range of tasks in a variety of environments may still be a distant reality. Breakthroughs are needed in multiple areas, including AI, machine learning, materials science, and control systems. The path to achieving such versatility is a gradual process, with robots gradually evolving from being task-specific to becoming more versatile and eventually becoming general-purpose.

Will household robots (beyond vacuum cleaners) become commonplace within the next 10 years?

Over the next decade, domestic robots are likely to emerge in certain areas such as elderly care and home security. But the vision of a general-purpose domestic robot that can autonomously perform a variety of household tasks is probably even further away. The challenges are not only technical, but also include aspects such as affordability, user acceptance, and ethical considerations.

What are some important robotics stories/trends that aren’t getting enough coverage?

Despite significant advances in specific niches and the success of robot implementation in specific industries, these stories are often overshadowed by the appeal of more futuristic or generic robot stories. It happens often. Incremental and impactful successes in areas such as agriculture, healthcare, or special industrial applications deserve further attention as they represent real and tangible progress in the field of robotics. .