Robot signatures in our behaviours

我们行为中的机器人签名

• 批准号: RGPIN-2020-05146
• 负责人: Moon, AJung
• 金额: $ 2.91万
• 依托单位: McGill University
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2022
• 资助国家: 加拿大
• 起止时间: 2022-01-01 至 2023-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=752204
• 关键词: Robot; signatures; our; behaviours

Humans are hard-wired to be influenced by our physical environment and behaviours of those around us. We habitually mimic postures of those we attend to, match our tone of voice and pace of speech during a conversation, and naturally synchronize our footsteps while walking side-by-side. Today, the factors that influence our behaviours increasingly include intelligent machines, such as robots that interact with us and share our physical space. Unlike the influence other people have on us, however, robots can influence people in ways that can be programmed, one-sided, deployed and at scale. What machine-induced influence on people is socially acceptable? What kind of machine-induced influence is harmful to individuals or considered undesirable manipulation? These are some of the many normative questions that must be answered with interdisciplinary means as we continue to design and deploy automation technologies to address some of the world's most challenging problems. The long-term vision for the proposed research program is to address the following question: "Given that the autonomous intelligent systems that interact with us inherently influence our decisions and actions, how can we design the systems to maximize the benefits while protecting the users from potential harm?" I aim to investigate the dynamics of influence robots have on people to enable (a) technologists to make appropriate decisions in the designing of interactive robotic systems and (b) policymakers to make evidence-based technology policy that pertains to the deployment of the systems in our society. In the next five years, I plan to take the first steps to tackle the research question by investigating how movements of a robot influence our decisions and actions. Two Ph.D. students, one Master's in Engineering (thesis) student, and five undergraduate summer interns will be hired under this interdisciplinary research program. Together, they will (a) develop a framework to quantify robot influence on our behaviours, (b) how robot motions influence the quality of our decisions, and (c) explore how what it means to design robot behaviours to be transparent about its effects to the users. Under my supervision, the students will be trained on a highly interdisciplinary set of techniques, ranging from real-time trajectory design, dynamical systems modelling, and information-theory to behavioural economics. The output from the proposed, interdisciplinary research will be crucial for (a) roboticists to be thoughtful in our design as we navigate the moral grey zone between robot behaviours that are beneficial and manipulative; (b) business leaders to manage the risks that can arise from deploying interactive robots in the industry; and (c) policymakers in shaping meaningful technology policies in Canada and abroad. It also has the potential to serve an integral role for roboticists in industry and academia by helping to monitor the impact of the products and services we provide.

人类天生就会受到周围环境和行为的影响，我们习惯性地模仿我们所关注的人的姿势，在谈话时配合我们的语气和说话的速度，并在走路时自然地同步我们的脚步。如今，影响我们行为的因素越来越多地包括智能机器，例如与我们互动并共享我们的物理空间的机器人，然而，与其他人对我们的影响不同，机器人可以以多种方式影响人类。被编程，什么样的机器对人产生的影响是社会可以接受的？什么样的机器产生的影响对个人有害或被认为是不良操纵？这些是必须通过跨学科来回答的一些规范问题。意味着我们继续设计和部署自动化技术来解决世界上一些最具挑战性的问题，拟议研究计划的长期愿景是解决以下问题：“鉴于与我们互动的自主智能系统本质上会产生影响。我们的决定和行动，我们如何设计系统的利益最大化，同时保护用户免受潜在伤害？”我的目标是研究机器人对人类影响的动态，以使（a）技术人员能够在设计交互式机器人系统时做出适当的决策，以及（b）政策制定者能够做出与我们社会中的系统部署有关的基于证据的技术政策 在接下来的五年中，我计划通过调查机器人的运动如何影响我们的决策和行动来解决这个研究问题。 .D. 学生，一名工程硕士（论文）学生和五名本科生暑期实习生将在这个跨学科研究计划中共同聘用，他们将（a）开发一个框架来量化机器人对我们行为的影响，（b）机器人运动如何影响我们决策的质量， (c) 探索如何设计对用户透明的机器人行为，在我的监督下，学生将接受一系列高度跨学科的技术培训，包括实时设计轨迹、动力系统。建模和信息论拟议的跨学科研究的成果对于（a）机器人专家在我们的设计中深思熟虑至关重要，因为我们在机器人行为之间的道德灰色地带中导航是有益的和操纵性的；（b）企业领导者管理风险； （c）政策制定者在制定加拿大和国外有意义的技术政策时，它还有可能通过帮助监测机器人的影响，为工业界和学术界的机器人专家发挥不可或缺的作用。产品和我们提供的服务。