Table Cleaning Task by Human Support Robot Using Deep Learning Technique

View full-text article in PMC

. 2020 Mar 18;20(6):1698. doi: 10.3390/s20061698

Algorithm 1 Cleaning Path Generation

1:
Step 1
2:
N is an array holds all the nodes position
3:
P is an array used to store the cleaning path
4:
$q_{s t a r t}$ = leftmost(N); Set uppermost left nodes as $q_{s t a r t}$
5:
load $q_{n}$ = $N - 1$
6:
$q_{g o a l}$ = def_goal(litter collection point); the function load the litter collection point as $q_{g o a l}$
7:
Step 2
8:
while cleaning path between $q_{s t a r t}$ to $q_{n}$ not discovered do
9:
Q, $b a c k t r a c k n o d e$ , $i n t e r n o d e s$ $a n d$ $P a t h$ are variable used for store the temporary value
10:
for $n = 0; n < q_{n}; n + +$ do
11:
Q= dfs_paths( $q_{s t a r t}$ , $q_{n}$ ); Explore the path between $q_{s t a r t}$ and $N - 1$ nodes
12:
end for
13:
$b a c k t r a c k n o d e$ = backtrack.find(Q); the function find the backtrack node in the explored path
14:
$i n t e r n o d e s$ = group.find( $q_{s t a r t}$ , $b a c k t r a c k n o d e$ ); the function collect intermediate node between $q_{s t a r t}$ and $b a c k t r a c k n o d e$
15:
Path = def_PRM_node_connect( $q_{s t a r t}$ , $i n t e r n o d e s$ , $b a c k t r a c k n o d e$ , $q_{g o a l}$ , R); PRM function connect the nodes according to threshold function R
16:
plt.plot ( $P a t h$ )
17:
end while
18:
visited.add( $P a t h$ ); Mark all path identified node in N
19:
Store the path in P
20:
load $q_{n}$ = $N - P a t h$ ; exclude path generate node from N
21:
set.new( $q_{s t a r t}, N$ ); set new $q_{s t a r t}$ from updated N
22:
Run the Step 2 up-to generate the path for remaining nodes