The issue of explanations for AI systems cooperating with humans has been a topic of considerable interest of late. But it is widely argued that current solutions that are based on local representations do not fully capture the reasoning behind the underlying decision.
University of Edinburgh
Referential communication refers to how an agent communicates their intended meaning to another agent about a referent, or object. This is a crucial aspect of human-robot interaction, especially when the robot must collaborate with human users in order to help them perform a shared task.
The aim of the ‘MiniMat’ project is to develop a minimal animal-like soft robot and test its impact on vulnerable children in cruelty prevention and animal-assisted interventions (AAI). Intentional animal cruelty begins as young as 4 years of age, often by highly vulnerable children. MiniMat is required because of the ethical issues of using live animals in interventions with children who have harmed animals. Similarly, animal-assisted interventions can benefit vulnerable children’s mental health.
In real world applications, a large variety of jobs involve handling various objects as the core process. However, most of these jobs nowadays are still performed by people. One obvious fact of human manipulation is the use of two hands that cooperate to achieve some task. But not too much research has been done involving coordination and optimal use of two hands. At the system level, having two hands naturally introduces physical redundancy that provides extra layer of safety and reliability against malfunctions such as object slippage.
Research Background of the Project: Soft manipulators have recently gained significant recognition for their versatility and advantages over rigid-link systems. Their multi-DOF bending capabilities and their intrinsic compliance make them a viable technology for complex tasks where an enhanced degree of flexibility is required as well as locomotion tools for bioinspired robots. However, the complexity of these continuum actuators makes them hard to control and, consequently, less suited for everyday use in industrial, medical or other robotics applications.
There is significant interest in characterising surgical skills in the form of detailed models that correlate hand movements, applied forces and the context of the shape and texture of tissues that are actually being manipulated.
Objects play a central role in the behaviour of intelligent systems like robots. This makes semantic object segmentation a fundamental basic component of many applications. State-of-the-art object segmentation is often done training expensive networks with lots of labelled images. This means that new knowledge is expensive to acquire, since the network needs lots of labelled images of each new object. In addition, these networks take as input only single images, ignoring all the useful information from the motion of objects.
The ignition of flammable liquids and gases in offshore oil and gas environments is a major risk and can cause loss of life, serious injury, and significant damage to infrastructure. Power supplies that are used to provide regulated voltages to drive motors, relays, and power electronic controls can produce heat and cause sparks. As a result, the European Union requires ATEX certification on electrical equipment to ensure safety in such extreme environments. Implementing designs that meet this standard is time-consuming and adds to the cost of operations.
Operations in hostile environments–such as those found in Nuclear Decommissioning, Oil and Gas, Mining, and Space–require the execution of sophisticated tasks. Examples of these tasks are: building structures, and deploying tools for inspection, lifting, and cutting. These environments pose a significant risk to the health and safety of manual workers. The safety of personnel can be mitigated by the use of autonomous robotic systems which can perform the required tasks in these extreme environments.
These environments present the following challenges: