Coffee with Finder (#ICRA2022)

As part of her role as one of the recipients of the IEEE ICRA 2022 Science Communication Awards, Avi Ravindran sat roughly with a few researchers from academia and industry who attended the conference. Curious about what they would say? Read their quotes below!

“I truly believe that the techniques gained, especially imitation and transfer learning, will enable scalable bot applications in human and unstructured environments. We are on the verge of seeing bot agents adapt and solve real-world problems dynamically”

– Nicholas Nadeau, Chief Technology Officer, Halodi Robotics

“On the one hand, I think the interaction between cognition and control is very exciting, in terms of common core principles, while on the other hand, it’s fascinating and inspiring to see more robots coming out of the lab.”

– Mathias Matamala, PhD student, Oxford Dynamic Robot Systems, Oxford Institute for Robotics

“I believe that incorporating introductions in relation to current landscape architecture and temporal consistency found in the context of mobile robotics, can be used to guide the learning of more robust representations”

– KAVISHA VIDANAPATHIRANA, QUT & CSIROR ROBOTS

“Right now, my goal is to find out what researchers need in order to care about their motivation and well-being”

– Daniel Carrillo-Zapata, founder of scientific agitation

“We have a huge amount of unsupervised knowledge and are always updating our introductions. Leveraging unsupervised prior training on a large scale and owning a lifelong learning system seems to be an important step in the right direction”

— Nitish Dashora, researcher, Berkeley AI Research & Redwood Center for Theoretical Neuroscience

“When objects are in disarray, with different objects on top of each other, the robot needs to interactively and independently rearrange the scene in order to retrieve the target body position with the fewest actions to achieve overall efficiency. I am working on position estimation algorithms to process dense visual data. as well as dispersed tactile data.”

– Prajval Kumar, BMW and University of Glasgow

“Thinking about why robots or even structures behave the way they do, and framing the questions and answering them in this line satisfies my curiosity as a researcher”

Tong Ta is a postdoctoral researcher at the University of Tokyo

“I sometimes hear leg movement is a problem that has been resolved, but I don’t agree. I think performance standards have just been raised and we can collectively handle more dynamic, efficient and reliable gaits”

– Kevin Green, Ph.D. Candidate, Oregon State University

“My goal in robotics research is to lower the cost and improve the capabilities of marine research platforms by introducing modularity and inefficiency in the field. We are working to understand how to bring our team swimming technology into today’s streaming environments”

Jedalia Knegnik, Ph.D. Candidate, GRASP Laboratory and Modular Robotics Laboratory, University of Pennsylvania

“I am interested in how we can develop the algorithms and representations needed to enable long-range robotic navigation without human intervention, such as in the case of an autonomous underwater robot constantly mapping out a marine ecosystem for an extended period of time. There are a lot of challenges such as how we can build a compact representation of the world, grounded ideally on human-understood semantics? How do we nimbly deal with cognition outliers that inevitably occur in a lifelong environment? Also, how do we extend methods for estimating a robot’s state in time and space while correlating memory and computational requirements?”

– Kevin Doherty, Computer Science and Artificial Intelligence Laboratory, MIT and Woods Hole Oceanographic Institutions

“How can robots learn to interact with themselves and engage with the world without an intuitive sense of either? Communication is at the core of biological and robotic systems. Inspired by control theory, information theory, and neuroscience, early work in artificial intelligence (AI) and robotics focused on a class of systems Dynamics known as feedback systems These systems are characterized by repetitive mechanisms or feedback loops that control, regulate or “orient” system behavior toward desired stable states in the presence of perturbation in diverse environments Feedback between sensation, prediction, decision, action, and return is an element Critical to the sensory learning needed to achieve robust intelligent robotic systems in the wild, a major challenge in the field Existing robots are fundamentally numb to the world, limiting their ability to sense themselves and their environment.This problem will increase as robots grow in complexity, dexterity and maneuverability, guided by Feedback control systems such as proportional integral derivative (PID), reinforcement learning (RL), and model predictive control (MPC) are now popular in robotics, as (optimal, Bayesian) Kálmán filtering for point-based IMU-GPS signals. The imperfect is the multimodal, high-dimensional distributed sensations necessary to perceive general intelligent behavior, and to carry out complex sequences of actions through high-level abstractions created from an intuitive sense or understanding of physics. Parallel Distributed Processing Engines (PDPs), most digital artificial neural networks are completely separated from the sensors and only provide a negative picture of the world. We are working to change that by combining parallel distributed sensing and data processing through a neural model. This includes innovations in hardware, software, and data sets. At Nervosys, we aim to make this dream a reality by building the first neural system and platform for general machine intelligence.”

– Adam Erickson, founder of Nervosys

Tags: c-events


Daniel Carrillo-Zapata received his Ph.D. in swarm robotics at the Bristol Robotics Laboratory in 2020. He is now promoting a culture of “scientific agitation” to engage in two-way conversations between researchers and the community.

Daniel Carrillo-Zapata received his Ph.D. in swarm robotics at the Bristol Robotics Laboratory in 2020. He is now promoting a culture of “scientific agitation” to engage in two-way conversations between researchers and the community.


Ahalia Ravindran is a PhD student at the Australian Center for Field Robotics, University of Sydney, Australia.

Ahalia Ravindran is a PhD student at the Australian Center for Field Robotics, University of Sydney, Australia.

Leave a Reply

Your email address will not be published.