Multiagent Systems Engineering
Intelligent Vehicle Systems

Two vehicles sharing information.

The Multiagent Systems Engineering (MaSE) approach is used to design physical and behavioral characteristics of individual components (agents) of a complex system. This becomes important when the individual components have semi- or fully-autonomous capabilities and it is desired to understand and control the behaviors of the overall system whose behaviors emerge from the interaction of the agents over time. These systems often display behaviors that are not exhibited by any individual agent and, to ensure system-level compliance, requires an understanding of how these behaviors emerge.

Key Concepts of Multiagent Systems Engineering

Some key concepts for MaSE are:

• Individuals do not contain global knowledge
• Individuals interact in specific ways within specific environments
• Feedback dynamics exist between individuals and overall system, where system behavior is an emergent property
• Agent-based modeling and simulation is required to understand the emergent system behaviors
  • Use results to refine physical or behavioral design of individuals

High-level diagram for situation awareness.

Engineering Application of Multiagent Systems

Intelligent Transportation Systems

Multiagent systems within intelligent transportation systems are more commonly referred to as Cooperative Vehicle Systems (CVS). In these systems, the individual agents can be mobile devices, such as vehicles, or static devices such, as Road-Side Equipment (RSE) installed on intersections. The U.S. Department of Transportation (USDOT) initiative to promote technologies for both mobile and static devices within a CVS is called Connected Vehicle. In contrast to a centrally planned traffic management system, none of the individual devices in a CVS need contain an understanding of global (system-wide) events or objectives.

Two vehicles share information about a pedestrian.

Some key concepts for CVS are:

• “Market penetration”
  • Perception-enhanced vehicles, infrastructure
  • Increasing communications among vehicles (V2V) and with infrastructure (VII)
• Cooperative vehicle behaviors
  • V2V sensor sharing for pedestrian protection
  • Infrastructure integration
  • Cooperative teaming for traffic smoothing

SwRI engineers developed MARTI^® for commercial applications, including convoy operations, using off-the-shelf components.

 

Vehicle convoy demonstration using three manned vehicles.

Manufacturing Systems

Vehicle convoy demonstration using MARTI autonomous vehicle.

Another application in which multiagent systems might be transferred to an engineered system is in manufacturing and warehousing logistics control design. These systems can be very complex in their organizational and operational structures and are traditionally controlled using top-down (hierarchical) techniques. This approach, however, introduces limits on the functional complexity and operational efficiency of the system. Hierarchical control also introduces operational fragility because the system is unable to adapt to anomalous or unforeseen events. In terms of a warehousing process, MaSE may be used to identify equipment and process requirements in terms of physical capabilities and behavioral modes.

Automated Guided Vehicles (AGVs) are widely used in automated warehousing functions; however, these machines operate at very low levels of autonomy – in highly constrained environments and under close monitoring. An AGV-like machine may be identified as part of a multiagent system analysis, but its behavioral modes would be such that the AGV could operate without external guidance and monitoring, in a highly unconstrained environment, able to adapt quickly to changes in its environment or goal. In this case, the AGV is an agent within the automated warehouse multiagent system. However, operators may also want to coordinate the behaviors of individual AGVs as a subsystem. The MaSE approach then seeks to determine the individual AGV characteristics and behaviors that will lead to a coordination of all AGVs toward a predefined system-level goal. For instance, the AGVs, as a group, will behave differently if they are able to actively communicate with each other, as opposed to sensing the activities of other AGVs through changes in the environment. The group will also behave differently if an individual AGV seeks out other AGVs, rather than avoids them, etc.

CS simulation results.

Graph showing team distance error. Click image for larger view.

Multiagent system supply chain.

Supply Chain Management

A common application of agent-based modeling and multiagent systems analysis is in the area of Supply Chain Management (SCM) and optimization. The individual pieces of a supply chain can include large manufacturing facilities, small overseas suppliers, and individuals who stock the shelves at a retailer. The precise movement of information and material is critical to the efficient operation of a supply chain. MaSE modeling and simulation can provide insight into critical supply chain issues, such as how many people, trucks, or buildings should be allocated for a specific function and how much inventory to maintain.

Related Terminology

agent-based systems  •  multiagent systems  •  distributed intelligence  •  decentralized control  •  architectures  •   multiagent systems engineering  •  engineered system