Multi-agent transformer-accelerated RL for satisfaction of STL specifications
One of the main challenges in multi-agent reinforcement learning is scalability as the number of agents increases. This issue is further exacerbated if the problem considered is temporally dependent. State-of-the-art solutions today mainly follow centralized training with decentralized execution paradigm in order to handle the scalability concerns. In this paper, we propose time-dependent multi-agent transformers which can solve the temporally dependent multi-agent problem efficiently with a centralized approach via the use of transformers that proficiently handle the large input. We highlight the efficacy of this method on two problems and use tools from statistics to verify the probability that the trajectories generated under the policy satisfy the task. The experiments show that our approach has superior performance against the literature baseline algorithms in both cases.
Code (0)
등록된 구현이 없습니다.
Tasks
Multi-agent Reinforcement LearningSimilar Papers 제목 키워드 기반
Risk-Aware Real-Time Task Allocation for Stochastic Multi-Agent Systems under STL Specifications
This paper addresses the control synthesis of heterogeneous stochastic linear multi-agent systems with real-time allocation of signal temporal logic (STL) specifications. Based on previous work, we decompose specificatio…
Autonomous DrivingModel Predictive ControlNeural Network-based Control for Multi-Agent Systems from Spatio-Temporal Specifications
We propose a framework for solving control synthesis problems for multi-agent networked systems required to satisfy spatio-temporal specifications. We use Spatio-Temporal Reach and Escape Logic (STREL) as a specification…
Automating the Refinement of Reinforcement Learning Specifications
Logical specifications have been shown to help reinforcement learning algorithms in achieving complex tasks. However, when a task is under-specified, agents might fail to learn useful policies. In this work, we explore t…
Reinforcement LearningSwarm-STL: A Framework for Motion Planning in Large-Scale, Multi-Swarm Systems
In multi-agent systems, signal temporal logic (STL) is widely used for path planning to accomplish complex objectives with formal safety guarantees. However, as the number of agents increases, existing approaches encount…
Computational EfficiencyMotion PlanningLeveraging Classification Metrics for Quantitative System-Level Analysis with Temporal Logic Specifications
In many autonomy applications, performance of perception algorithms is important for effective planning and control. In this paper, we introduce a framework for computing the probability of satisfaction of formal system …
Multi-class Classification