研究目标
针对飞行员对AI赋能机载座舱智能系统的信任程度难以量化的问题，面向典型有/无人协同任务场景开展人机信任影响机理、演化机制和定量模型研究。通过研究人机自主决策能力边界随任务、态势环境变化的特征，建立信任动态演化模型，提出人机权限分配原则，并开发可计算的信任判别方法。旨在构建基于信任的人机权限分配原则，提升人机协同任务的效能和安全性，最终为机载座舱智能系统的设计和应用提供理论基础和技术支持。
研究成果
Conference Articles
-
A Human-Machine Interaction Approach for Overtaking Scenarios Based on Reinforcement Learning and Intent Inference
Huan Wang,
Yuwen Gan,
and Yunbo Zhao
In The 44th Chinese Control Conference (CCC 2025)
2025
[Abs]
[pdf]
In the context of overtaking maneuvers within hybrid human-machine driving scenarios in autonomous driving, traditional approaches predominantly focus on directly modeling the system’s dynamic behavior and subsequently framing it as an optimal control problem. However, these methods are heavily reliant on the accuracy of the models employed, which often contain inherent errors. Furthermore, given the significant variability among individual human drivers and the influence of their unobservable internal states on their actions, control strategies that are solely based on mathematical models are inadequate for predicting fluctuations in human driving strategies. This limitation can lead to huge fluctuations in control volumes. To solve the above problem, this paper adopts an approach that unites a reinforcement learning module(PPO) with an intent inference module.This approach allows for the acquisition of effective decision-making strategies through interaction with the environment, thereby circumventing the need for modeling unknown entities. Additionally, the application of Bayesian inference for intent inference abstracts the determinants of human behavior as intentions, thereby enhancing the predictive capabilities of machines regarding human decision-making and facilitating more efficient and smoother control. The efficacy of this proposed method is validated through simulation experiments.
Journal Articles
-
Enhancing Human–Machine Collaboration: A Trust-Aware Trajectory Planning Framework for Assistive Aerial Teleoperation
Qianzheng Zhuang,
Kangjie Huang,
Xiaoran Jin,
Pengfei Li,
Yunbo Zhao,
and Yu Kang
Machines
2025
[Abs]
[doi]
[pdf]
Human–machine collaboration in assistive aerial teleoperation is frequently compromised by trust imbalances, which arise from the vehicle’s complex dynamics and the operator’s constrained perceptual feedback. We introduce a novel framework that enhances collaboration by dynamically integrating a model of human trust into the unmanned aerial vehicle’s trajectory planning. We first propose a Machine-Performance-Dependent trust model, specifically tailored for aerial teleoperation, that quantifies trust based on real-time safety and visibility metrics. This model then informs a trust-aware trajectory planning algorithm, which generates smooth and adaptive trajectories that continuously align with the operator’s trust level and intent inferred from control inputs. Extensive simulations conducted in diverse forest environments validate our approach. The results demonstrate that our method achieves task efficiency comparable to that of a trust-unaware baseline while significantly reducing operator workload and improving trajectory smoothness, achieving reductions of up to 23.2% and 43.2%, respectively, in challenging dense environments. By embedding trust dynamics directly into the trajectory optimization loop, this work pioneers a more intuitive, efficient, and resilient paradigm for assistive aerial teleoperation.
项目人员
赵云波 东昊楠 卢峻森 夏睿钰 欧阳晨 武昱泽 汪洋 甘毓雯 金骁然 黄康杰
项目合作