Distributed coordination of a formation of heterogeneous agents with individual regrets and asynchronous communications

Application of Collective Movement in Real Life Scenarios: Overview of Current Flocking Solutions

“…In addition to the table it can be concluded that all the examined solutions (except Carlési's [11]) use an infinite communication medium. Therefore, there is no need for a medium access protocol since there is no packet losses and each node can transmit messages at any time regardless of the medium usage.…”

“…In the previous models the communication graph (where vertices are nodes and edges indicate the communication links among them) was undirected, moreover, in the simplest model the whole graph was complete, [42] passive N/A N/A Navarro's [34,1] passive, active velocity and weights 3.3m Ç elikkanat's [13] passive, active heading vector ∼ 20m Cavagna's [12] active position ∞ Hai-Tao's [63] active position r, a distance limit Olfati-Saber's [36] active position r, a distance limit Zhang's [62] active position, velocity r, a distance limit Carlési's [11] active velocity and control variable ∞ and in the second case it is composed of many smaller (the size is determined by the interaction range of the device) complete graphs. A realistic communication environment cannot be described as an undirected graph, instead some works have adopted a directed graph, since the transmission ranges can differ from node to node.…”

“…In a more realistic scenario, the blind flood protocol cannot be efficient anymore, since the radio channels (for instance in an urban environment) can be heavily loaded and the periodically initiated broadcasts can cause many collisions and even broadcast storms. In [11] communication is supposed to be limited, possibly subject to random failures, and asynchronous. Two different scenarios have been identified: a low and a high rate scenario.…”

“…10 times in a second) and with the periodically received messages the control model can calculate different type of parameters from the received data (such as velocity form the position or acceleration from the velocity) with relatively high accuracy. Only the framework proposed by Carlési et al [11] triestime slot to tackle the problems of communication failures, but still many idealistic assumptions are used for the communication environment, such as the radio range being infinite. The scalability of the network is usually not examined: when the flock gets bigger (consisting of more and more nodes), the number of the transmitted messages in a single timeslot increases as well, but since the communicational environment is almost perfect, the messages will not collide.…”

“…The scalability of the network is usually not examined: when the flock gets bigger (consisting of more and more nodes), the number of the transmitted messages in a single timeslot increases as well, but since the communicational environment is almost perfect, the messages will not collide. The almost perfect communicational environment is assumed in some of the algorithms [11,51], however an error is added to the received data, but it is utilised at control layer and not at the communication layer.…”