Odor Source Localization in Obstacle Regions Using Switching Planning Algorithms with a Switching Framework

. 2023 Jan 19;23(3):1140. doi: 10.3390/s23031140

Algorithm 1 Algorithms switching framework

Data: agent, source ▹ Location of agent and odor source

gridMap, obstacle ▹ Given environmental components

1:
procedureOSL
2:
$sourceFound \leftarrow$ false
3:
for all $c_{i j}$ in $g r i d M a p$ do
4:
$p (c_{i j}) \leftarrow 1 /$ size( $g r i d M a p$ ) ▹ Initialize uniform distribution
5:
end for
6:
$L_{e n v} \leftarrow$ decompose( $g r i d M a p, o b s t a c l e$ ) ▹ $L_{e n v} = {E_{1}, E_{2}, \dots, E_{K}}$
7:
while sourceFound = false do
8:
$E_{k^{*}} \leftarrow$ max( $L_{e n v}$ )
9:
if $a g e n t$ in $E_{k^{*}}$ then ▹ Switch to Infotaxis algorithm
10:
$a c t i o n \leftarrow \underset{a \in A^{'}}{argmax} E [Δ S (a)]$
11:
$a g e n t \leftarrow$ move(agent, action)
12:
else ▹ Switch to Dijkstra algorithm
13:
$j u m p T a r g e t \leftarrow \underset{c_{i j} \in E_{k^{*}}}{argmax} p (c_{i j})$
14:
$p a t h \leftarrow$ dijkstra(agent, jumpTarget)
15:
$a g e n t \leftarrow$ move(agent, path)
16:
end if
17:
$s e n s o r V a l u e \leftarrow$ sampling()
18:
$g r i d M a p \leftarrow$ update( $a g e n t, s e n s o r V a l u e, g r i d M a p$ )
19:
if $∥ a g e n t - s o u r c e ∥ < targetRadii$ then
20:
$s o u r c e F o u n d \leftarrow$ true.
21:
end if
22:
end while
23:
end procedure