Heuristic Evaluation of Swarm Metrics' Effectiveness

Abstract

Typical visualizations of robot swarms (greater than 50 entities) display each individual entity; however, it is immensely difficult to maintain accurate position information for each member in real-world situations with limited communications. Generally, it will be difficult for humans to maintain an awareness of all individual entities. Further, the swarm's tasks may impact the desired visualization. Thus, an open question is how best to visualize a swarm given various swarm tasks. This paper presents a heuristic evaluation that analyzes the application of swarm metrics to different swarm visualizations and tasks. A brief overview of the visualizations is provided, along with a description of the heuristic metrics and the analysis.

Human Interaction With Robot Swarms: A Survey

“…Reliable metrics regarding the performance of a swarm in real applications are not available, apart from some early results in [141] and [142], and their general need is identified in several swarm surveys [5]- [7]. From the perspective of HSI, this is a promising area of research, and when monitoring swarms, operators can greatly benefit from such metrics, e.g., when managing competing tasks while monitoring the quality of service in the communication network.…”

Two Invariants of Human Swarm Interaction

“…The evidence for supporting this was mostly indirect, meaning that we never directly measured how humans perceived a swarm as opposed to, say, a decoherent set of randomly-moving independent agents. As mentioned in the section on related literature, recent results have made attempts to directly measure how a human perceives a swarm; initial results, though encouraging, still need further validation (Harriott et al, 2014;Manning et al, 2015). Trends in the recent work provide further support for the claim of perception of collective unit.…”

“…Work by Cummings (2004) has investigated supervisory control of swarms of UAVs; however, this work differs from our work because the systems used are not described by the dynamical models that are critical to our invariants. Recent results have also made attempts to directly measure how a human perceives a swarm; initial results, though encouraging, still need further validation (Harriott, Seiffert, Hayes, & Adams, 2014;Manning, Harriott, Hayes, Adams, & Seiffert, 2015;.…”

Swarm Transparency

“…A more cohesive representation of the overall swarm progress is needed. Abstract visualizations [12] may support level 1 transparency, but it is unclear exactly what information needs to be communicated to humans' in order to support transparency.…”