Corridor with obstacle#
In a serie of two notebooks, we look at a simple but more interesting scenario than Empty environment.
In the first notebook, we explore the scenario defined as
type: CorridorWithObstacle
length: 1.0
width: 1.0
min_radius: 0.2
max_radius: 0.2
groups:
-
type: thymio
number: 1
radius: 0.08
control_period: 0.05
color: firebrick
kinematics:
type: 2WDiff
wheel_axis: 0.094
max_speed: 0.12
behavior:
type: HL
optimal_speed: 0.12
horizon: 10
tau: 0.25
eta: 0.5
safety_margin: 0.05
barrier_angle: 1.0
state_estimation:
type: Bounded
range: 1.0
update_static_obstacles: true
where one agent wants to travel along a corridor with constant speed, avoiding the single obtacle. We also get familiar with the inputs/outputs spaces of the models, which we train in the second notebook, to navigate using a combination of two sensors
type: Combination
sensors:
- type: Boundary
min_y: 0
max_y: 1
range: 1
- type: Discs
number: 1
range: 1
max_speed: 0.0
max_radius: 0.0
include_valid: false
to detect the obstacle and the corridor walls.