Homeland safety and security will increasingly depend upon autonomous unmanned vehicles as a method of assessing and maintaining situational awareness. As autonomous team algorithms evolve toward requiring less human intervention, it may be that having an “operator-in-the-loop” becomes the ultimate goal in utilizing autonomous teams for surveillance. However studies have shown that trust plays a factor in how effectively an operator can work with autonomous teammates. In this work, we study mechanisms that look at autonomy as a system and not as the sum of individual actions. First, we conjecture that if the operator understands how the team autonomy is designed that the user would better trust that the system will contribute to the overall goal. Second, we focus on algorithm input criteria as being linked to operator perception and trust. We focus on adding a time-varying spatial projection of areas in the ROI that have been unseen for more than a set duration (STEC). Studies utilize a custom test bed that allows users to interact with a surveillance team to find a target in the region of interest. Results show that while algorithm training had an adverse effect, projecting salient team/surveillance state had a statistically significant impact on trust and did not negatively affect workload or performance. This result may point at a mechanism for improving trust through visualizing states as used in the autonomous algorithm.

Physical interference between robots happens when robots try to occupy the same region. In multi-robot experiments, interference between robots has been found to negatively impact performance. In addition, simulations do not always accurately predict real robot performance. So, we study the effects of unmodeled factors, such as interference, on the difference between real and simulated experiments. We believe that it is beneficial to understand the categorization of robot interference. Therefore, we examine different forms of interference in real robot experiments in order to set a basis for an interference model in simulation. We show that interference can be broken into categories which have discrete characteristics that affect robot performance differently in various environments.

In cooperative multirobot systems, communication can speed up completion, reduce redundancy, and prevent interference between robots. Typically, wireless point-to-point communication is used to coordinate robots. However, environmental interference, unpredictable network conditions, and distances between robots can affect the reliability of wireless communication. Therefore, approaches other than continuous message passing throughout exploration are useful. We consider the problem of coordinating a multirobot system to explore an unknown, large, open environment. An approach that uses sector search with rendezvous is presented. Robots explore an environment in sectors, or designated areas, and periodically meet to communicate map information of what they have explored. Our approach is compared to other communication paradigms in simulation. Results suggest that sector search with rendezvous is more efficient than having no communications. It further demonstrates advantages over scenarios in which robots communicate only with other robots in close proximity, and is comparable to a role-based approach with dynamic team hierarchies.

Simulations are typically used to model a problem and find a solution before real world testing. They speed up the validation process and allow researchers to modify their code accordingly. However, a problem occurs when simulation results are not consistent with real world results. Researchers have found inconsistencies due to odometry error and team size. However, no research has studied the effects specific to robot teams that affect the realism of multi-robot experiments. This paper shows how simulation results vary from experimental results when conducting multi-robot experiments. Simulation and real experiments are performed using different environments and cooperation paradigms. Results show that specific environmental features and cooperation paradigms significantly affect the usefulness of simulated results when predicting performance of real robot teams.

We consider the problem of dispersing nodes of a mobile sensor network to cover an unknown environment. Approaches that use observation to infer state and intent to disperse robot nodes are presented. Nodes use their observations to decide on their next actions. With simulated and physical node experiments, we compare observation-based approaches to no communications, direct communications, and potential field approaches. Experimental results show that using observation to infer state and intent to disperse nodes performs better than non-communicative and potential field based cooperation and in some cases direct communications.

In this paper, we propose an extension to existing mobile robot development environments that explicitly declares the frequency requirements for client controller threads. This extension enables better use of robot resources by running modules only as fast as needed. Developers are forced to consider the frequency requirements and their impacts, which should result in better code. Physical experiments employed a K-team Koala robot. Preliminary results showing the effect of explicitly defining frequency are also presented.