This article proposes a novel stochastic-skill-level-based shared control framework to assist human novices emulate experts in complex dynamic tasks. The proposed aims infer stochastic-skill-levels (SSLs) of the and provide personalized assistance based on inferred SSLs. SSL can be assessed as stochastic variable that denotes probability novice will behave similarly experts. We propose data-driven method characterize demonstrations model expert an model, respectively. Then, our inference...

In this paper, a hybrid shared controller is proposed for assisting human novice users to emulate expert within human-automation interaction framework. This work motivated let learn the skills of using automation as medium. Automation interacts with in two folds: it learns how optimally control system from experts demonstrations by offline computation, and assists real time without excess amount intervention based on inference novice's skill-level our properly designed controller. takes more...

Autonomous systems are increasingly being used for the purpose of training humans to attain new skills or perform tasks. In these contexts, autonomous should be responsive to, and guide, human behavior such that skill task performance is maximized. These generally rely on determine if assistance needed. However, it recognized also respond cognitive factors, as self–confidence, relevant learning. We propose experimentally validate a heuristic control strategy, based both user's self-reported...

In this paper, we present a Model-Free Stochastic Inverse Optimal Control (IOC) algorithm for the discrete-time infinite-horizon stochastic linear quadratic regulator (LQR). Our proposed exploits richness of available system trajectories to recover control gain <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$K$ </tex-math></inline-formula> and cost function parameters notation="LaTeX">$(Q,R)$ in low...

In this paper, we propose a human-automation interaction scheme to improve the task performance of novice human users with different skill levels. The proposed includes two modes: learn from experts mode and assist novices mode. mode, automation learns expert user such that awareness objective is obtained. Based on learned objective, in customizes its control parameter towards emulating user. We experimentally test designed quadrotor simulation environment, results show approach capable...

Human behavior modeling is important for the design and implementation of human-automation interactive control systems. In this context, human refers to a human's input We propose novel method that uses demonstrations given task infer unknown objective variability. The represents intent or desire. It can be inferred by inverse optimal improve understanding providing an explainable function behind behavior. Meanwhile, variability denotes intrinsic uncertainty in described Gaussian mixture...

We propose a data-driven forward stochastic reachability analysis algorithm for Human-In-The-Loop (HITL) systems. focus on certain type of HITL system whose behavior is dominated by human operator, example, multi-rotor controlled operator. In such system, the intervention operator may generate conservative reachable set due to unpredictable control strategy The proposed computes less accounting operator's behavior, i.e., we present that considers unknown controller information system....

Reachability analysis is a widely used method to analyze the safety of Human-in-the-Loop Cyber Physical System (HiLCPS). It allows HiLCPS respond against an imminent threat in advance by predicting reachable states system. However, it could lead unnecessarily conservative set if prediction only relies on system dynamics without explicitly considering human behavior, and thus risk might be overestimated. To avoid conservativeness, we present state probability distribution function (pdf) which...

This paper proposes a quantitative framework for optimally allocating task functions in human-autonomy teaming (HAT). HAT involves cooperation between humans and autonomous agents to achieve common goals. As possess different capabilities, function allocation plays crucial role ensuring effective HAT. However, designing the best adaptive remains challenge, as existing methods often rely on qualitative rules intensive human-subject studies. To address this limitation, we propose computational...

예쁜 꼬마선충은 모델 생물로서 지금까지 행동과 이를 제어하는 신경세포들 사이의 관계를 밝히기 위한 많은 연구들이 수행되었다. 본 연구에서는 표면의 강성이 다른 고체 환경에서 꼬마선충의 운동관련 적응행동을 연구하였다. 위에서 움직일 때 기는 파형을 조절함으로써 기계적으로 환경에 적응을 한다. 즉, 외부환경이 더 단단해질수록 파형의 진폭과 파장이 감소하게 된다. 흥미로운 사실은 기계적인 감각에 결함이 있는 돌연변이의 경우 정상 꼬마선충과는 보인다는 것이다. 이것은 효과적으로 적응하기 위해서 자극을 감지하고 반응하고 적응하는 기작이 있음을 의미한다. 이에 꼬마선충이 과정을 설명할 수 신경회로 모델을 제안하였다.

Reachability analysis is a widely used method to analyze the safety of Human-in-the-Loop Cyber Physical System (HiLCPS). This strategy allows HiLCPS respond against an imminent threat in advance by predicting reachable states system. However, it could lead unnecessarily conservative set if prediction only relies on system dynamics without explicitly considering human behavior, and thus risk might be overestimated. To reduce conservativeness reachability analysis, we present state which takes...

Autonomous systems are increasingly being used to assist humans in learning skills of growing complexity. To achieve success accelerating a human’s learning, autonomous should be responsive to, and guide, human behavior such that task performance is maximized. Existing systems, as intelligent tutoring (ITS), typically rely on the feedback drives their decision-making [1] . However, it has been recognized these respond cognitive involved relationship between user [2] For example, an ITS...

This letter presents a novel inverse optimal control (IOC) approach that can account for uncertainties in measurements and system models. The proposed IOC aims to recover an objective function including time-varying term, called variability, from given demonstration. All of the demonstration model be lumped into variability such optimality condition violation is further reduced. inferred has two advantages over by existing approaches: first, enhance capability describing since it represents...