**4. Swarm principles**

The Swarm computing approach can be better understood by examining its two underlying principles: resource sharing and autonomy. These principles can be subdivided into more specific aspects that guide the Swarm evolution, as shown in **Figure 1**.

Autonomy confers devices to the ability to share resources without needing manual input from humans (self-organization). These automatic interactions are not programmed into the system beforehand but emerge naturally during execution time (spontaneity). As a result, the participants may encounter optimized paths and *Swarm Computing: The Emergence of a Collective Artificial Intelligence at the Edge… DOI: http://dx.doi.org/10.5772/intechopen.110907*

#### **Figure 1.**

*The Swarm principles and their subsumed aspects.*

risky interactions over time, from which they must learn to unexpected situations. Autonomy can be further subdivided into the following aspects:


Resource sharing allows a set of entities (distribution) to provide and consume resources from each other for mutual benefit (cooperation). These entities may have different amounts (heterogeneity) of resources to share, use open and interoperable protocols (openness), and cooperate independently of their physical structure and location (dematerialized). Finally, resources can be composed at different levels (granular), and their scope may range from local to global (scalable). The following aspects are key to enabling resource sharing in the Swarm:


by anyone. Despite, is that, security, and privacy risks, it makes systems more powerful (efficient, through sharing of resources), more resilient (by the use of redundant resources through dynamic reconfiguration), and more capable, enabling applications that have not been realized yet;


The Swarm principles act as guidelines for the research and development of the Swarm computing paradigm and its applications. For example, imagine a set of cameras and smart doors that cooperate. When a person approaches the system, the cameras will tell the doors whether to open or not. Or in an emergency scenario, a car passing by accident can automatically share a danger alert with surrounding devices. Both examples work without human intervention (autonomy) and create value through cooperation (resource sharing). By combining these two principles in a cohesive model/platform/abstraction, the Swarm creates a new era for the Internet of Things.
