Class Meeting 01: Welcome & Grand Challenges in Robotics

Today's Class Meeting

Welcome to and overview of Intro Robotics (here is a link to my slides)
Small group discussions on 10 grand challenges in robotics

Grand Challenges in Robotics

Today, we'll be holding a series of small group discussions centered around 10 grand challenges of robotics identified in the paper: Yang, Guang-Zhong, et al. The grand challenges of Science Robotics. Science robotics 3.14 (2018): eaar7650.

Logistics:

You can find the discussion breakout room assignments in this Google Spreadsheet (you have to be signed in with your @uchicago.edu email address to view the spreadsheet).
As you discuss these grand challenges, use this Google Jamboard - posting a stickey note for each point your group discusses and any relevant images or drawings.
For each small group discussion you engage with, before your discussion ends, identify one idea that your group discussed that you will share with the entire class at the conclusion of this activity.

In the text that follows, I've provided some background information (often quotations from the Science Robotics article) and suggested discussion questions to help your group get started with each of the 10 grand challenges.

1) New Materials and Fabrication Schemes

"Gears, motors, and electromechanical actuators are fundamental to many of the robotic platforms in use today, but laboratories around the world have begun to explore new materials including artificial muscles, compliant materials for soft robots, and emerging advanced manufacturing and assembly strategies. As illustrated in Fig. 2 [the figure below], these promise a new generation of robots that are power-efficient, multifunctional, compliant, and autonomous in ways that are similar to biological organisms." Additional capabilities of robots using new materials could include "the robot building and repairing itself."

Questions to spark discussion:

What kinds of new robot morphologies can you imagine might be developed using new materials such as artificial muscles, compliant materials for soft robots, and emerging advanced manufacturing and assembly strategies?
If robots can be built with materials other than gears, motors, and electromechanical actuators, what additional capabilities do you think robots will be able to have?
What do you think it would take in order for a robot to be able to rebuild itself?
What applications do you think would benefit most from robots made out of new materials and fabrication strategies?

2) Biohybrid and Bioinspired Robots

Bioinsipred robotics: "the use of fundamental biological principles that are translated into engineering design rules for creating a robot that performs like a natural system."

Biohybrid robotics: "the direct use of biological material to design synthetic machines" based on "biological understanding."

"Specific grand challenge lists for biorobotics have remained largely unchanged for the past 30 years — a battery to match metabolic conversion, muscle-like actuators, self-healing material that manufactures itself, autonomy in any environment, human-like perception, and, ultimately, computation and reasoning."

Questions to spark discussion:

Like birds and other flying organisms inspired the design of airplanes, what are other biological systems that may be able to inspire robot designs?
Can you imagine any biohybrid robots that people may find useful? (e.g., Hiroshi Ishii's group at the MIT media lab recently created a "living textile" where bacteria actuated the material to help dancers regulate heat)
What can we learn from evolution about the design of useful and viable systems?
What capabilities (e.g., running, jumping, digging, swimming, throwing) of living biological systems do you think are the most important for the robots of today to develop? What biological systems could inspire or enable these capabilities?

3) Power and Energy

"As for any electronic system, power and energy sources represent one of the most challenging areas of robotics research and deployment, especially for mobile robotics. Underwater and particularly deep-sea exploration requires compact, stable, high–energy density batteries to support robots working in challenging conditions and extreme environments. The increasing adoption of drones and autonomous vehicles is fueling the development of new battery technologies that are safe and affordable, with longer cycle lives, robust temperature tolerance, higher energy densities, and relatively low weight [...] Fundamental issues being addressed remain the same for many historical technologies: irreversible phase transitions of active materials and/or unstable electrode-electrolyte solution interfaces."

Questions to spark discussion:

In what ways do you think wireless robots might be able to "extract useful energy from their surroundings"?
If we were able to solve some of the problems of power and energy, what do you imagine robots would be able to do without today's limitations?
What applications and industries would be the most affected with the development of more light weight, temperature tolerant, high energy density, and affordable battery technologies?

4) Robot Swarms

"Robot swarms allow simpler, less expensive, modular robotic units to be reconfigured into a team depending on the task at hand while being as effective as a larger, task-specific, monolithic robot, which may be more expensive and have to be rebuilt depending on the task. Nature provides a repertoire of examples that illustrate this idea. Independently acting organisms cannot achieve a goal by themselves but, in coordination with other organisms, can solve complex problems and complete a mission [...] The swarm principle can be used at macro-, micro-, and nanoscales with a plethora of application areas."

Questions to spark discussion:

For what tasks and applications do you think that re-configurable robot swarms would be most beneficial?
What emergent behavior can you imagine from a swarm of robots that are each programmed in a simple way? (e.g., an individual fish might move in a 'simple' way, but together with many other fish, the behavior of a school of fish emerges)
What are advantages of swarms? Why would someone choose to purchase a robot swarm instead of a single larger robot?

5) Navigation and Exploration

"Path planning, obstacle avoidance, localization, and environment mapping are ubiquitous requirements of robot navigation and exploration [...] Many robots we deploy are intrinsic explorers that we send to the far reaches of the planet — the deep oceans, under the Arctic ice pack, into volcanoes — and go where no human has yet tread, often under unknown and extreme conditions. The associated challenges are therefore much greater than those encountered today."

Questions to spark discussion:

How do you think robots might be able to overcome lapses in communication during navigation? (e.g., robotic spacecraft may experience "long latency and low bandwidths of communications not only greatly reduce productivity but also put the survival of the robot itself at risk")
Current mapping and navigation techniques assume that the world is static and does not change. In what contexts do you think it might be most important to enable robots to be able to model "time-varying, dynamic environments with deformable objects"?
A large challenge of navigation is handling "failures and being able to adapt, learn, and recover." What capabilities do you think robots will need to have in order to recover and learn from failures?

6) AI for Robotics

"The advent of deep learning methods resulted in remarkable levels of object recognition accuracy using hierarchical pattern recognition that retained information coherence at each level of the hierarchy. The new machine-learning algorithms were combined with unprecedented access to data, as well as inexpensive and powerful computing hardware."

"However, we still have a long way to go to replicate and exceed all the facets of intelligence that we see in humans [...] Meta-learning, or learning how to learn new things, is a critical new AI capability not only with large training data sets but also with limited data. The challenge is to be able to learn on the fly, adapting to dynamic and uncertain environments

Questions to spark discussion:

What are some recent machine learning developments or algorithms you've heard about?
What problems are machine learning and deep learning algorithms good at solving? Which problems might they be bad at solving?
What application domains for robots do you think might benefit the most from robots that are able to "learn how to learn new things"?
What do you think a robot would have to know in order to "learn on the fly"?

7) Brain-Computer Interfaces

"A BCI forges a direct, online communication between brain and machine, independent from the user’s physical abilities, and represents a new way to augment human capabilities and restore patient function [...] BCIs translate the user’s intentions into outputs or actions by means of machine-learning techniques, operating by either presenting a stimulus to the operator and waiting for his/her response (synchronous) or continuously monitoring the operator’s cognitive activity and responding accordingly (asynchronous)."

Questions to spark discussion:

What applications do you think brain-computer interfaces (BCI) would be most useful for?
Alternatives to BCI include eye tracking and muscle-based devices, what advantages do you think BCI have compared with these other approaches?
BCI can be both invasive and noninvasive? What do you think are the benefits and drawbacks of invasive and noninvasive BCI?

8) Social Interaction

"Robotics and AI have often underestimated the difficulty of replicating capabilities that humans find particularly easy [...] Because humans are so adept at recognizing and interpreting social behavior, we often underestimate the complexity of the challenge that this represents for a robot [...] The three most significant challenges that stem from building robots that interact socially with people are modeling social dynamics, learning social and moral norms, and building a robotic theory of mind."

Questions to spark discussion:

If you were to design a robot behavior to signal to a human waitress at a restaurant (filled with humans) that the robot is ready for the bill, how would you do it?
What social and moral norms do you think are the most important for robots to adhere to in order to act appropriately in human society? (An example of a social norm is being quiet when you're in a library. An example of a moral norm is the view that murder, under most circumstances, is wrong.)
"Social interaction also requires building and maintaining complex models of people, including their knowledge, beliefs, goals, desires, and emotions." What aspects do you think a robot might need to model when trying to predict people's intentions? (e.g., predicting which item in a clothing store a person wants to buy, predicting whether someone is about to cross a street)

9) Medical Robots

"The impact of robotics on medicine is undeniable. The therapeutic and commercial success of Intuitive Surgical’s da Vinci system has spurred a number of commercial ventures targeting surgical applications, which echo the emerging trend in precision surgery, focusing on early malignancies with minimally invasive intervention and greater consideration of patient recovery and quality of life [...] One of the primary challenges in surgical and interventional robotics is a move toward systems that exhibit increasingly higher degrees of autonomy. A second grand challenge is the creation of fully implantable robots that replace, restore, or enhance physiological processes. A third grand challenge is in the realization of micro- and nanorobotic devices of clinical relevance."

Questions to spark discussion:

In what ways can robots augment what humans can do in medical settings? Said another way, what can robots do better than people in medical settings?
One useful application of robots in the medical domain may be a robot that could help a patient before an ambulance or human help can arrive. A robot could "assess the condition of a patient quickly, prioritize problems, and often take time-urgent steps to stabilize the patient." If a robot is given this level of autonomy and responsibility, who would be the responsible party if the robot malfunctions or inadvertently kills a patient? How might we avoid these type of outcomes?
"Perhaps the most significant challenge of automating any clinical task is to be able to anticipate, detect, and respond to all possible failure modes." If robots are given the job of performing tasks, like large portions of surgeries, on their own, how should possible failures be anticipated and handled?

10) Robot Ethics and Security

"With increasing levels of autonomy and human-robot interaction, there needs to be careful consideration of potential regulatory, ethical, and legal barriers and the context of how robots are deployed. Because robotics and AI are fueled by data, some challenges are rooted in human-environment interactions and data governance, especially consent, discrimination, fairness, ownership, privacy, surveillance, and trust."

Questions to spark discussion:

"Excessive reliance on robotics and AI may lead to the delegation of sensitive tasks to autonomous systems that should remain at least partly subject to human supervision." Let's take, as an example, autonomous vehicles (AVs). How might over-reliance on AVs be problematic? And how would you design an AV to include human supervision?
Robots and AI have the potential to replace many human jobs. Do you have any ideas on how we can both 1) leverage the benefits of robots and AI and 2) keep people employed?
Robots and AI may be programmed with biases or, as a result of learning from humans, may 'learn' to be biased. How do you think we can prevent and limit bias in AI and robotics applications?

Additional Resources

Rodney Brooks (my PhD advisor grandfather -- my PhD advisor's PhD advisor) posted predictions about the future of artificial intelligence (AI) and machine learning (ML) in 2018, as well as and an annual update on those predictions every year afterwards. After today's discussion you may these predictions about the future of AI and ML an interesting read.