forked from gtianyi/motionPlanning
-
Notifications
You must be signed in to change notification settings - Fork 0
/
problemSetting.inst
74 lines (46 loc) · 2.06 KB
/
problemSetting.inst
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
#This is how long the planner is allowed to run
Timeout ? 60
#This is how much memory the planner is allowed to use (MB)
Memory ? 1000
#This is the seed used to initialize the random number generator
Seed ? 1
#This is how many runs will be executed -- the way I have things set up it'll always be 1
Runs ? 1
#This is the filename where you want the results to be stored
Output ? outfile
#This is an OMPL parameter for using intermediate states when propagating, you'll almost always want this to be true
AddIntermediateStates ? true
#Some of the planners accept a goal bias parameter, in this case we've specified RRT which uses this paramter
GoalBias ? 0.05
#This is the propagation step size used when forward simulating the system, it has a serious impact on collisions if set to high
PropagationStepSize ? 0.05
#This is basically, for each propagation what is the minimum and maximum number of unit propagations to do
MinControlDuration ? 1
MaxControlDuration ? 100
#If there is not a controller that can be used for steering, we generate this many random controls and take the one that gets closest to the target
NumControls ? 10
#This is the motion model we're using
Domain ? DynamicCar
#This is the agent mesh used to represent the agent (for collision checking)
AgentMesh ? car2_planar_robot.dae
#This is the mesh used for the environment
EnvironmentMesh ? forest.dae
#In a lot of our planners we need environment bounds to create an abstraction
EnvironmentBounds ? -30 30 -30 30
#The kinematic car is a planar vehicle, so we set a start x,y for it (we don't play with the orientation but it can be done)
Start ? 3 -5
#The goal location
Goal ? 0 -25
#In our experiments we are focused on getting spatially close to the goal so this signifies the radius in x,y or x,y,z around the goal that is acceptable
GoalRadius ? 1
#Finally the planner to be used
Planner ? BEAST
#Beast which search
WhichSearch ? D*
StateRadius ? 6
PRMSize ? 1000
NumEdges ? 5
ValidEdgeDistributionAlpha ? 10
ValidEdgeDistributionBeta ? 1
InvalidEdgeDistributionAlpha ? 1
InvalidEdgeDistributionBeta ? 10