Trust-Based Human-Robot Collaboration (HRC) for High-Level Distributed Multi-Robot Motion Planning with Temporal Logic Constraints
Abstract
The research problem of this YIP grant is to synthesize symbolic motion plans (SMP) for multi-robot systems (MRS) with a human-in-the-loop through quantitative trust analysis. The objective is to develop scalable and adaptive multi-robot motion plans based on trust analysis for improved performance of the joint human-robot collaboration (HRC) systems. A dynamic, quantitative, and probabilistic human-to-robot trust model is developed by combining causal reasoning and direct evidence. This model also factors in the correlation among neighboring robots. Compositional reasoning approaches are developed to decompose the global task specification. An optimal multi-robot task allocation approach based on iterative automata decomposition has been developed. Our approach works with general task specifications and considers the task allocation problem with heterogeneous robots. Trust is used as a metric for specification decomposition and updated in a dynamic fashion. A trust-based runtime monitoring and switching mechanism is proposed for tradeoffs between task safety and efficiency. Deadlock- and livelock-free algorithms are designed to guarantee reachability of goals with a human-in-the-loop. Both ROS Gazebo and Matlab simulations are utilized for robot symbolic motion planning with a human-in-the-loop.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 04, 2019
- Accession Number
- AD1104377
Entities
People
- Yue Wang
Organizations
- Clemson University