1,720,988 research outputs found
RAi UK Annual Report 2024-25
No full textWe estimate that RAi UK has secured at least £3 million in ‘follow on funding’ from other research grants and activities. We have created this publication to help communicate and accelerate the impact of our pioneering research programme. We hope it will be a starting point for further engagements with stakeholders, who will make use of the work in a responsible and considered way. We will continue to collaborate and embed many of the lessons learnt across the ecosystem
The effect of automated agents on individual performance under induced stress
No full textInduced stress is a phenomenon commonly experienced across different fields such as emergency services, healthcare, air traffic control, sports, and business - which necessitates the development of effective coping strategies and resilience for individuals or teams performing under pressure. This study aims to examine the effects of automated agents on individual performance during high-stress conditions. The design of these agents ensures they carry out identical tasks as participants based on predetermined frameworks. Participants underwent an experimentally designed task that aimed at inducing stress while measuring their performance amidst time pressure and auditory distraction. Results indicate that working with automated agents causes individuals to alter their approach by focusing narrowly on immediate concerns - making it challenging for them to consider several options or see broader contexts accurately. Regardless of ability level participants' performances were influenced by these automated agents. Future research will explore how these findings interact with physiological signals. This study highlights the importance of developing effective coping strategies and the potential impact of social factors on individual performance under induced stress
CrowdAR: a live video annotation tool for rapid mapping
No full textDigital Humanitarians are a powerful and effective resource to analyse the vast amounts of data that disasters generate. Aerial vehicles are increasingly being used for gathering high resolution imagery of affected areas, but require a lot of effort to effectively analyse, typically taking days to complete. We introduce CrowdAR, a real-time crowdsourcing platform that tags live footage from aerial vehicles flown during disasters. CrowdAR enables the analysis of footage within minutes, can rapidly plot snippets of the video onto a map, and can reduce the cognitive load of pilots by augmenting their live video feed with crowd annotations
Planning search and rescue missions for UAV teams
No full textThe coordination of multiple Unmanned Aerial Vehicles (UAVs) to carry out aerial surveys is a major challenge for emergency responders. In particular, UAVs have to fly over kilometre-scale areas while trying to discover casualties as quickly as possible. To aid in this process, it is desirable to exploit the increasing availability of data about a disaster from sources such as crowd reports, satellite remote sensing, or manned reconnaissance. In particular, such information can be a valuable resource to drive the planning of UAV flight paths over a space in order to discover people who are in danger. However challenges of computational tractability remain when planning over the very large action spaces that result. To overcome these, we introduce the survivor discovery problem and present as our solution, the first example of a continuous factored coordinated Monte Carlo tree search algorithm. Our evaluation against state of the art benchmarks show that our algorithm, Co-CMCTS, is able to localise more casualties faster than standard approaches by 7% or more on simulations with real-world data
Beyond text: multi-modal LLM in human robot interaction
No full textMultimodal interaction plays a vital role in Human-Robot Interaction (HRI), enabling robots to communicate with humans through multiple channels. This study introduces a novel approach to enhance such interactions by treating images and human motion as distinct foreign languages, in addition to text. In the proposed framework, vector quantization is employed to convert multimodal inputs such as images and human motions to an aligned set of tokens. A Large Language Model (LLM) is then pre-trained with the use of Low-Rank Adaptation (LoRA) and instruction-tuned on a dialogue dataset that incorporates both image and motion context. The proposed multimodal LLM framework aims to equip robots with the ability to understand and respond to complex human queries through multimodal inputs and outputs, enabling more natural and effective interactions
Evaluating international AI skills policy: a systematic review of AI skills policy in seven countries
No full textAs artificial intelligence (AI) is having an increasingly disruptive impact across industries, companies continue to report having difficulty when recruiting for AI roles, while new graduates find it difficult to find employment, indicating a skills gap or skills misalignment. International approaches to AI skills programmes can offer a guide to future policy development of a skilled workforce, best placed to harness the economic opportunities that AI may support. The authors performed a systematic literature review on AI skills in government policies and documents from seven countries: Australia, Canada, China, Singapore, Sweden, the United Kingom and the United States. We found a divide between countries which emphasised a broader, nationwide approach to upskill and educate all citizens at different levels, namely the United States and Singapore and those countries which emphasised a narrower focus on educating a smaller group of experts with advanced AI knowledge and skills, namely China, Sweden and Canada. We found that the former, broader approaches tended to correlate with higher AI readiness and index scores than the narrower, expert-driven approach. Our findings indicate that, to match world-leading AI readiness, future AI skills policy should follow these broad, nationwide approaches to upskill and educate all citizens at different levels of AI expertise.</p
Interactive scheduling of appliance usage in the home
No full textWe address the problem of recommending an appliance usage schedule to the homeowner which balances between maximising total savings and maintaining sufficient user convenience. An important challenge within this problem is how to elicit the user preferences with low intrusiveness, in order to identify new schedules with high cost savings, that still lies within the user’s comfort zone. To tackle this problem we propose iDR, an interactive system for generating personalised appliance usage scheduling recommendations that maximise savings and convenience with minimal intrusiveness. In particular, our system learns when to stop interacting with the user during the preference elicitation process, in order to keep the bother cost (e.g., the amount of time the user spends, or the cognitive cost of interacting) minimal. We demonstrate through extensive empirical evaluation on real–world data that our approach improves savings by up to 35%, while maintaining a significantly lower bother cost, compared to state-of the-art benchmarks
AI management essentials (AIME) consultation response
No full textWe are submitting this response to the Department for Science, Innovation and Technology (DSIT) consultation on the new AI Management Essentials (AIME) tool1 on behalf of Responsible AI UK (RAi UK), an open and multidisciplinary network that brings together researchers from across the four nations of the UK to understand how we should shape the development of AI to benefit people, communities and society. To arrive at this response, we sent out a call to Principal Investigators and Co-Investigators of RAi UK funded projects2 to contribute to this consultation. What follows is a synthesis of the responses we received from our research community. Overall, RAi UK sees the AIME tool as a valuable first step for Small to Medium Sized Enterprises (SMEs) and Startups to adopt towards implementing robust and responsible AI governance practices. Our contributions below aim to improve the tool’s usability and impact
Factored Monte-Carlo tree search for coordinating UAVs in disaster response
No full textThe coordination of multiple Unmanned Aerial Vehicles (UAVs) to carry out surveys is a major challenge for emergency responders. In particular, UAVs have to fly over kilometre-scale areas while trying to discover casualties as quickly as possible. However, an increase in the availability of real-time data about a disaster from sources such as crowd reports or satellites presents a valuable source of information to drive the planning of UAV flight paths over a space in order to discover people who are in danger. Nevertheless challenges remain when planning over the very large action spaces that result. To this end, we introduce the survivor discovery problem and present as our solution, the first example of a factored coordinated Monte Carlo tree search algorithm to perform decentralised path planning for multiple coordinated UAVs. Our evaluation against standard benchmarks show that our algorithm, Co-MCTS, is able to find more casualties faster than standard approaches by 10% or more on simulations with real-world data from the 2010 Haiti earthquake
Trade-offs of dynamic control structure in human-swarm systems
No full textSwarm robotics is a study of simple robots that exhibit complex behaviour only by interacting locally with other robots and their environment. The control in swarm robotics is mainly distributed whereas centralised control is widely used in other fields of robotics. Centralised and decentralised control strategies both pose a unique set of benefits and drawbacks for the control of multi-robot systems. While decentralised systems are more scalable and resilient, they are less efficient compared to the centralised systems and they lead to excessive data transmissions to the human operators causing cognitive overload. We examine the trade-offs of each of these approaches in a human-swarm system to perform an environmental monitoring task and propose a flexible hybrid approach, which combines elements of hierarchical and decentralised systems. We find that a flexible hybrid system can outperform a centralised system (in our environmental monitoring task by 19.2%) while reducing the number of messages sent to a human operator (here by 23.1%). We conclude that establishing centralisation for a system is not always optimal for performance and that utilising aspects of centralised and decentralised systems can keep the swarm from hindering its performance
- …
