Robotics seminars

Humans are natural teachers; learning through instruction is one of the most fundamental ways that we learn. Interactive Task Learning (ITL) is an emerging research agenda that studies the design of complex intelligent robots that can acquire new knowledge through natural human teacher-robot learner interactions. ITL methods are particularly useful for designing intelligent robots whose behavior can be adapted by humans collaborating with them. In this talk, I will summarize our recent findings on the structure that human instruction naturally has and motivate an intelligent system design that can exploit their structure. The system – AILEEN – is being developed using the common model of cognition. Architectures that implement the Common Model of Cognition - Soar, ACT-R, and Sigma - have a prominent place in research on cognitive modeling as well as on designing complex intelligent agents. However, they miss a critical piece of intelligent behavior – analogical reasoning and generalization. I will introduce a new memory – concept memory – that integrates with a common model of cognition architecture and supports ITL.

Cognitive science and cognitive neuroscience researchers have agreed that the replication of findings is important for establishing which ideas (or theories) are integral to the study of cognition across the lifespan. Recently, high-profile papers have called into question findings that were once thought to be unassailable. Much attention has been paid to how p-hacking, publication bias, and sample size are responsible for failed replications. However, much less attention has been paid to the fidelity by which researchers enact study protocols. Researchers conducting education or clinical trials are aware of the importance in fidelity – or the extent to which the protocols are delivered in the same way across participants. Nevertheless, this idea has not been applied to cognitive contexts. This seminar discusses factors that impact the replicability of findings alongside recent models suggesting that even small fidelity deviations have real impacts on the data collected.

Socially Assistive Robots (SARs) have shown great potential to help children in therapeutic and healthcare contexts. SARs have been used for companionship, learning enhancement, social and communication skills rehabilitation for children with special needs (e.g., autism), and mood improvement. Robots can be used as novel tools to assess and rehabilitate children’s communication skills and mental well-being by providing affordable and accessible therapeutic and mental health services. In this talk, I will present the various studies I have conducted during my PhD and at the Cambridge Affective Intelligence and Robotics Lab to explore how robots can help assess and rehabilitate children’s communication skills and mental well-being. More specifically, I will provide both quantitative and qualitative results and findings from (i) an exploratory study with children with autism and global developmental disorders to investigate the use of intelligent personal assistants in therapy; (ii) an empirical study involving children with and without language disorders interacting with a physical robot, a virtual agent, and a human counterpart to assess their linguistic skills; (iii) an 8-week longitudinal study involving children with autism and language disorders who interacted either with a physical or a virtual robot to rehabilitate their linguistic skills; and (iv) an empirical study to aid the assessment of mental well-being in children. These findings can inform and help the child-robot interaction community design and develop new adaptive robots to help assess and rehabilitate linguistic skills and mental well-being in children.

How could machines learn as efficiently as humans and animals? How could machines learn to reason and plan? How could machines learn representations of percepts and action plans at multiple levels of abstraction, enabling them to reason, predict, and plan at multiple time horizons? I will propose a possible path towards autonomous intelligent agents, based on a new modular cognitive architecture and a somewhat new self supervised training paradigm. The centerpiece of the proposed architecture is a configurable predictive world model that allows the agent to plan. Behavior and learning are driven by a set of differentiable intrinsic cost functions. The world model uses a new type of energy-based model architecture called H-JEPA (Hierarchical Joint Embedding Predictive Architecture). H-JEPA learns hierarchical abstract representations of the world that are simultaneously maximally informative and maximally predictable.

It is common for animals to use self-generated movements to actively sense the surrounding environment. For instance, rodents rhythmically move their whiskers to explore the space close to their body. The mouse whisker system has become a standard model to study active sensing and sensorimotor integration through feedback loops. In this work, we developed a bioinspired spiking neural network model of the sensorimotor peripheral whisker system, modelling trigeminal ganglion, trigeminal nuclei, facial nuclei, and central pattern generator neuronal populations. This network was embedded in a virtual mouse robot, exploiting the Neurorobotics Platform, a simulation platform offering a virtual environment to develop and test robots driven by brain-inspired controllers. Eventually, the peripheral whisker system was properly connected to an adaptive cerebellar network controller. The whole system was able to drive active whisking with learning capability, matching neural correlates of behaviour experimentally recorded in mice.

TBC

Natural systems provide both puzzles to unravel and demonstrations of what's possible. The natural world is full of complex systems of dynamically interchangeable, individually unreliable components that produce effective and reliable outcomes at the group level. A complementary goal to understanding the operation of such systems is that of being able to engineer artifacts that work in a similar way. One notable type of collective behavior is collective construction, epitomized by mound-building termites, which build towering, intricate mounds through the joint activity of millions of independent and limited insects. The artificial counterpart would be swarms of robots designed to build human-relevant structures. I will discuss work on both aspects of the problem, including studies of cues that individual termite workers use to help direct their actions and coordinate colony activity, and development of robot systems that build user-specified structures despite limited information and unpredictable variability in the process. These examples illustrate principles used by the insects and show how they can be applied in systems we create.

As tiny robots become individually more sophisticated, and larger robots easier to mass produce, a breakdown of conventional disciplinary silos is enabling swarm engineering to be adopted across scales and applications, from nanomedicine to treat cancer, to cm-sized robots for large-scale environmental monitoring or intralogistics. This convergence of capabilities is facilitating the transfer of lessons learned from one scale to the other. Cm-sized robots that work in the 1000s may operate in a way similar to reaction-diffusion systems at the nanoscale, while sophisticated microrobots may have individual capabilities that allow them to achieve swarm behaviour reminiscent of larger robots with memory, computation, and communication. Although the physics of these systems are fundamentally different, much of their emergent swarm behaviours can be abstracted to their ability to move and react to their local environment. This presents an opportunity to build a unified framework for the engineering of swarms across scales that makes use of machine learning to automatically discover suitable agent designs and behaviours, digital twins to seamlessly move between the digital and physical world, and user studies to explore how to make swarms safe and trustworthy. Such a framework would push the envelope of swarm capabilities, towards making swarms for people.

Designing artificially intelligent systems and interfaces with socio-emotional skills is a challenging task. Progress in industry and developments in academia provide us a positive outlook, however, the artificial social and emotional intelligence of the current technology is still limited. My lab’s research has been pushing the state of the art in a wide spectrum of research topics in this area, including the design and creation of new datasets; novel feature representations and learning algorithms for sensing and understanding human nonverbal behaviours in solo, dyadic and group settings; designing longitudinal human-robot interaction studies for wellbeing; and investigating how to mitigate the bias that creeps into these systems. In this talk, I will present some of my research team’s explorations in these areas including social appropriateness of robot actions, virtual reality based cognitive training with affective adaptation, and bias and fairness in data-driven emotionally intelligent systems.

Multistable structures can reversibly change between multiple stable configurations when a sufficient energetic input is provided. While originally the field focused on understanding what governs the snapping, more recently it has been shown that these systems also provide a powerful platform to design a wide range of smart structures. In this talk, I will first show that pressure-deployable origami structures characterized by two stable configurations provide opportunities for a new generation of large-scale inflatable structures that lock in place after deployment and provide a robust enclosure through their rigid faces. Then, I will demonstrate that the propagation of transition waves in a bistable one-dimensional linkage can be exploited as a robust mechanism to realize structures that can be quickly deployed. Finally, while in the first two examples multistability is harnessed to realize deployable architectures, I will demonstrate that bistable building blocks can also be exploited to design crawling and jumping robots. Unlike previously proposed robots that require complex input control of multiple actuators, a simple, slow input signal suffices to make our system move, as all features required for locomotion are embedded into the architecture of the building blocks.

From slithering snakes, to entangling robots, self-deforming (shape changing) active systems display surprising dynamics. This is particularly true when such systems interact with environments or other agents to generate self-propulsion (movement). In this talk, I will discuss a few projects from my group illustrating unexpected effects in individual and collectives of self-deformers. For example, snakes and snake-like robots mechanically “diffract” from fixed environmental heterogeneities, collections of smart-active robots (smarticles) can locomote (and phototax) as a collective despite individual immobility, and geometrically actively entangling ensembles of blackworms and robots can self-propel as a unit to thermo or phototax without centralized control.

Non-traumatic brain injury due to stroke, cerebral palsy and HIV often result in serious long-term disability worldwide, affecting more than 150 million persons globally; with the majority of persons living in low and middle income countries. These diseases often result in varying levels of motor and cognitive impairment due to brain injury which then affects the person’s ability to complete activities of daily living and fully participate in society. Increasingly advanced technologies are being used to support identification, diagnosis, assessment, and therapy for patients with brain injury. Specifically, robot and mechatronic systems can provide patients, physicians and rehabilitation clinical providers with additional support to care for and improve the quality of life of children and adults with motor and cognitive impairment. This talk will provide a brief introduction to the area of rehabilitation robotics and, via case studies, illustrate how computer/technology-assisted rehabilitation systems can be developed and used to assess motor and cognitive impairment, detect early evidence of functional impairment, and augment therapy in high and low-resource settings.

Deep learning is a prime example of how brain-inspired computing can benefit development of artificial intelligence. But what else can we learn from the brain for bringing AI and robotics to the next level? Energy efficiency and data efficiency are the major features of the brain and human cognition that today’s deep learning has yet to deliver. The brain can be seen as a multi-agent system of heterogeneous learners using different representations and algorithms. The flexible use of reactive, model-free control and model-based “mental simulation” appears to be the basis for computational and data efficiency of the brain. How the brain efficiently acquires and flexibly combines prediction and control modules is a major open problem in neuroscience and its solution should help developments of more flexible and autonomous AI and robotics.

Henrik I. Christensen examines the scientific and engineering challenges involved in building intelligent robots that assist people. Advances in materials, embedded systems, sensors and machine learning support new combinations of perception, planning and actuation. The lecture considers the models underlying robot perception, planning and control, and how these components must be integrated into useful systems. Applications include customized manufacturing, transportation, security, support for older people and people with disabilities, and safer work for first responders in dangerous environments. Christensen discusses major research trends and the potential influence of these systems on work, home and leisure.