Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Safe Human-UAS Collaboration Abstraction

Hossein Rastgoftar

TL;DR

This work addresses safe collaboration between humans and a UAS in shared workspaces by integrating Petri Net abstractions of task evolution with a non-stationary Markov Decision Process for motion planning under human-intent uncertainty. The approach includes on-board perception to learn human intention and quantify distraction via a Moving Neighboring Set around human trajectories, and a time-varying cost for the UAS that is updated with real-time observations. Key contributions are the Petri Net constructs for WS transitions, the formalization of distraction and intention with probabilistic models, and the non-stationary MDP framework that yields safe, goal-oriented UAS trajectories in the presence of non-cooperative co-workers. The method demonstrates safe planning in simulations with multiple humans, offering a principled pathway to semi-autonomous co-working drones in complex environments.

Abstract

This paper studies the problem of safe humanuncrewed aerial system (UAS) collaboration in a shared work environment. By considering human and UAS as co-workers, we use Petri Nets to abstractly model evolution of shared tasks assigned to human and UAS co-workers. Particularly, the Petri Nets places represent work stations; therefore, the Petri Nets transitions can formally specify displacements between the work stations. The first objective is to incorporate uncertainty regarding the intentions of human co-workers into motion planning for UAS, when UAS co-workers closely interact with human co-workers. To this end, the proposed Petri Nets model uses conflict constructs to represent situations at which UAS deals with incomplete knowledge about human co-worker intention. The second objective is then to plan the motion of the UAS in a resilient and safe manner, in the presence of non-cooperative human co-workers. In order to achieve this objective, UAS equipped with onboard perception and decision-making capabilities are able to, through real-time processing of in-situ observation, predict human intention, quantify human distraction, and apply a non-stationary Markov Decision model to safely plan UAS motion in the presence of uncertainty.

Safe Human-UAS Collaboration Abstraction

TL;DR

This work addresses safe collaboration between humans and a UAS in shared workspaces by integrating Petri Net abstractions of task evolution with a non-stationary Markov Decision Process for motion planning under human-intent uncertainty. The approach includes on-board perception to learn human intention and quantify distraction via a Moving Neighboring Set around human trajectories, and a time-varying cost for the UAS that is updated with real-time observations. Key contributions are the Petri Net constructs for WS transitions, the formalization of distraction and intention with probabilistic models, and the non-stationary MDP framework that yields safe, goal-oriented UAS trajectories in the presence of non-cooperative co-workers. The method demonstrates safe planning in simulations with multiple humans, offering a principled pathway to semi-autonomous co-working drones in complex environments.

Abstract

This paper studies the problem of safe humanuncrewed aerial system (UAS) collaboration in a shared work environment. By considering human and UAS as co-workers, we use Petri Nets to abstractly model evolution of shared tasks assigned to human and UAS co-workers. Particularly, the Petri Nets places represent work stations; therefore, the Petri Nets transitions can formally specify displacements between the work stations. The first objective is to incorporate uncertainty regarding the intentions of human co-workers into motion planning for UAS, when UAS co-workers closely interact with human co-workers. To this end, the proposed Petri Nets model uses conflict constructs to represent situations at which UAS deals with incomplete knowledge about human co-worker intention. The second objective is then to plan the motion of the UAS in a resilient and safe manner, in the presence of non-cooperative human co-workers. In order to achieve this objective, UAS equipped with onboard perception and decision-making capabilities are able to, through real-time processing of in-situ observation, predict human intention, quantify human distraction, and apply a non-stationary Markov Decision model to safely plan UAS motion in the presence of uncertainty.
Paper Structure (14 sections, 28 equations, 6 figures, 2 algorithms)

This paper contains 14 sections, 28 equations, 6 figures, 2 algorithms.

Table of Contents

  1. Introduction
  2. Related Work
  3. Contributions
  4. Outline
  5. Preliminaries
  6. Problem Formulation
  7. Approach
  8. Problem 1: Distraction Probability Quantification
  9. Problem 2: Human Intention Prediction
  10. Problem 3: UAS Motion Planning
  11. MDP Components
  12. Solution
  13. Simulation Results
  14. Conclusion

Figures (6)

  • Figure 1: The constructs used for modeling human UAS collaboration.
  • Figure 2: The Petri Nets demonstrating human UAS interaction in a shared workplace.
  • Figure 3: Desired trajectories between the origin WS and three possible next WS are shown by dashed plot for human co-workers $1$, $2$, $3$, $4$, and $5$ in sub-figures (a), (b), (c), (d), and (e), respectively. Also, actual trajectory of each co-worker is shown by black dots in sub-figures (a)-(e).
  • Figure 4: Intention prediction of human co-workers.
  • Figure 5: Optimal actions taken by the UAS to safely plan its trajectory.
  • ...and 1 more figures

Theorems & Definitions (7)

  • Definition 1
  • Definition 2
  • Definition 3
  • Definition 4
  • Definition 5
  • Definition 6
  • Definition 7