Distributed Dynamic Event-Triggered Control for Multi-Agent Systems

Xinlei Yi,Kun Liu,Dimos V. Dimarogonas,Karl H. Johansson

DOI: https://doi.org/10.1109/CDC.2017.8264666

2017-04-19

Abstract:We propose two distributed dynamic triggering laws to solve the consensus problem for multi-agent systems with event-triggered control. Compared with existing triggering laws, the proposed triggering laws involve internal dynamic variables which play an essential role to guarantee that the triggering time sequence does not exhibit Zeno behavior. Some existing triggering laws are special cases of our dynamic triggering laws. Under the condition that the underlying graph is undirected and connected, it is proven that the proposed dynamic triggering laws together with the event-triggered control make the state of each agent converges exponentially to the average of the agents' initial states. Numerical simulations illustrate the effectiveness of the theoretical results and show that the dynamic triggering laws lead to reduction of actuation updates and inter-agent communications.

Optimization and Control

What problem does this paper attempt to address?

This paper aims to solve the consensus problem in multi - agent systems, especially by using event - triggered control (ETC) to achieve this goal. Specifically, the paper proposes two distributed dynamic triggering mechanisms to solve the consensus problem in multi - agent systems. Compared with the existing triggering mechanisms, these new dynamic triggering mechanisms introduce internal dynamic variables, which play a crucial role in ensuring that the triggering time sequence does not exhibit Zeno behavior. Zeno behavior refers to the existence of an infinite number of triggers within a finite time interval, which is a problem that needs to be avoided in multi - agent systems.

### Main Contributions
1. **Propose two distributed dynamic triggering mechanisms**: These two mechanisms do not require any prior knowledge of global network parameters.
2. **Prove that the proposed dynamic triggering mechanisms can make the multi - agent system reach consensus exponentially**.
3. **Prove that these dynamic triggering mechanisms can eliminate Zeno behavior**.
4. **Show that some existing triggering mechanisms are special cases of the control law proposed in this paper**.

### Background
- **Multi - agent consensus problem**: Multiple agents (such as mobile robots, autonomous underwater vehicles, unmanned aerial vehicles, etc.) need to reach an agreement on a quantity of interest (for example, the average value of their initial states).
- **Event - triggered control**: Different from time - driven sampling, event - triggered control only performs control updates and communications between agents when certain specific events occur (for example, when the state error exceeds a certain threshold).
- **Zeno behavior**: In continuous - time multi - agent systems, Zeno behavior means that there are an infinite number of triggers within a finite time interval, which needs to be avoided when designing triggering mechanisms.

### Methods
- **Static triggering mechanism**: The paper first discusses a static triggering mechanism, but this mechanism may not be able to completely eliminate Zeno behavior.
- **Dynamic triggering mechanism**: In order to explicitly eliminate Zeno behavior, the paper proposes two dynamic triggering mechanisms. These mechanisms introduce internal dynamic variables, thereby ensuring that the triggering time sequence does not exhibit Zeno behavior.

### Results
- **Simulation results**: The validity of the theoretical results is verified through numerical simulations, showing that the proposed dynamic triggering mechanisms can not only make the multi - agent system reach consensus, but also reduce the frequency of control updates and communications between agents.
- **No Zeno behavior**: The simulation results also show that the dynamic triggering mechanisms can effectively eliminate Zeno behavior.

### Conclusion
The dynamic triggering mechanisms proposed in this paper provide an effective solution to the consensus problem in multi - agent systems, especially in terms of eliminating Zeno behavior. Future research directions include considering more general linear multi - agent systems and dynamic self - triggering control.