Motion algorithm for autonomous rescue agents based on information assistance system

Search Algorithm for Teams of Heterogeneous Agents with Coverage Guarantees

“…It has been addressed as typical exploration missions by Monekosso and Remagnino [1] and Dollarhide et al [2]. It also fits the model of search and rescue applications as shown by Kurabayashi et al [3], Kantor et al [4], and Jennings et al [5]. A survey of classical searching techniques by Benkoski et al [6] provides additional background and references.…”

“…set ofz values defining center of occupancy based map cells B R cell centers reachable by agent in d steps d prediction horizon, number of waypoints in path d i minimum distance from generator i to setB max f χ ( ) reward function for course deviation in (℘ 2 ) f d ( ) reward function for distance in (℘ 2 ) f h ( ) reward function for high score cell in (℘ 2 ) h sensor reliability factor in range [0, 1] I index of agent closer toB max than any other agent In (m) indices corresponding to runs where at least m agents find the target J 0 ( ) total reward function for (℘ 2 ) LUM, VAGNERS, AND RYSDYK k counter for time index L x , L y width and height of map cell in x-direction and y-direction, respectively N x , N y number of columns and rows, respectively, of occupancy based map N(i) number of agents who find target in run ĩ N ave average number of agents who find target n(m) number of runs where at least m agents find the target P set of Voronoi generator points (℘ 1 ) subproblem of creating future world state estimates (℘ 2 ) subproblem of finding desirable cells for agent to search (℘ 3 ) subproblem of finding trajectories for agent's path p (A|B) conditional probability of A given B p i Voronoi generator point i q (a, b) calculates absolute angular difference between angles a and b R max maximum distance agent can travel in a d steps S (i, k) cumulative map score at step k for run i S ave (k) average cumulative map score at step k S max (k) maximum cumulative map score at step k S min (k) minimum cumulative map score at step k s k score of given occupancy map cell at step k (p(X k = x B ) at step k) T m (i) time when the mth agent finds the target for run i T m,ave average time when the mth agent finds the target V Voronoi diagram V max maximum velocity of agent V t variance of weights at time t V (i, k) variance of map scores at step k for run i V ave (k) average variance of map scores at step k V (z agt i ) Voronoi polygon associated with generator pointz agt i x spatial coordinate x A , x B event of target not in cell and target in cell states, respectively x min , x max minimum and maximum x value of occupancy based map x w (k,z) actual state of the world at time step k and locationẑ x w (k,z) estimated world state at time step k and locationz y min , y max minimum and maximum y value of the occupancy based map z (x, y) coordinate (P E , P N ) T z most desirable location in (℘ 2 ) z agt agent's current (x, y) position z A , z B observation of target not in cell and in cell, respectivelȳ z 0 agent's current coordinatē z H location of cell center with highest score in map and closest to agent z h location of cell center in agent's reachable set closest toz H z s location of cell center of same cell that agent is currently in…”