Update doc of dynamic random obstacles

changelog.md

@@ -44,7 +44,7 @@
 
 ## ir-sim 2.2.0
 
-- Rename the module from ir_sim to be irsim, rename the package name from ir_sim to be ir-sim
+- Rename the module from ir_sim to irsim, rename the package name from ir_sim to ir-sim
 - Add citation for the project
 - Refine the comments for the functions
 

doc/source/index.rst

@@ -50,3 +50,15 @@ The simple demonstrations of the simulator are shown below:
    :caption: API Documentation:
 
    api/modules
+
+
+Academic Cases
+--------------
+
+- `rl-rvo-nav(RAL & ICRA2023) <https://github.com/hanruihua/rl_rvo_nav>`_
+
+- `RDA_planner(RAL & IROS2023) <https://github.com/hanruihua/RDA_planner>`_
+
+Code Repository
+---------------
+`IR-SIM <https://github.com/hanruihua/ir-sim>`_

doc/source/usage/configure_behavior.md

@@ -66,7 +66,7 @@ The demonstration of the `rvo` behavior is shown in the following figure:
 - **`accer`:** The acceleration of the object.
 - **`factor`:** The factor to adjust the collision penalty. 
 
-Full list of behavior parameters can be found in the [YAML Configuration](../get_started/configuration/).
+Full list of behavior parameters can be found in the [YAML Configuration](../yaml_config/configuration/).
 
 
 ## Advanced Configuration for Custom Behavior

doc/source/usage/configure_dynamic_random_env.md

@@ -6,7 +6,83 @@ The dynamic, random, and clutter environment is useful to test the navigation an
 
 ## Random Obstacles Configuration Parameters
 
+Python script:
+
+```python
+import irsim
+
+env = irsim.make()   
+
+for i in range(1000):
+
+    env.step()
+    env.render(0.05)
+
+    if env.done():
+        break
+
+env.end()
+```
+
+YAML File:
 
 ```yaml
+world:
+  height: 50  # the height of the world
+  width: 50   # the height of the world
+  control_mode: 'keyboard'  
+
+robot:
+  - kinematics: {name: 'acker'} 
+    shape: {name: 'rectangle', length: 4.6, width: 1.6, wheelbase: 3}
+    state: [5, 5, 0, 0]
+    goal: [40, 40, 0]
+    vel_max: [4, 1]
+
+    sensors: 
+      - type: 'lidar2d'
+        range_min: 0
+        range_max: 20
+        angle_range: 3.14
+        number: 100
+
+    plot:
+      show_trajectory: True
+
+obstacle:
+
+  - number: 20
+    kinematics: {name: 'omni'}
+    distribution: {name: 'random', range_low: [10, 10, -3.14], range_high: [40, 40, 3.14]}
+    behavior: {name: 'rvo', wander: True, range_low: [10, 10, -3.14], range_high: [40, 40, 3.14], vxmax: 2, vymax: 2, factor: 3.0}
+    vel_max: [4, 4]
+    vel_min: [-4, -4]
+    shape:
+      - {name: 'circle', radius: 1.0, random_shape: True}  
+      - {name: 'polygon', random_shape: true, avg_radius_range: [0.5, 2.0], irregularity_range: [0, 0.4], spikeyness_range: [0, 0.4], num_vertices_range: [4, 6]}
+```
+
+The demonstration is shown in the following figure:
+
+```{image} gif/random_obstacles.gif
+:alt: random_obstacles
+:width: 400px
+:align: center
+```
+
+## Important Parameters Explained
+
+The parameters to generate random obstacles with various shapes are the settings of `behavior`, `distribution`, and `shape` in the `obstacle` section.
+
+- For the `rvo` behavior, set the `wander` to `True` to enable the random movement of the obstacles when they reach the goal position. And the `rvo` behavior is used to avoid the collision among obstacles. `range_low` and `range_high` are the lower and upper bounds of the random distribution of the goal position of the obstacles. 
+
+- The `distribution` parameter is used to set the random distribution of the obstacles in a certain area. The `range_low` and `range_high` are the lower and upper bounds of the random distribution of the initial position of the obstacles.
+
+- The `shape` parameter is used to set the random shape of the obstacles by setting the `random_shape` to `True`. For circular obstacles, the `radius` will be randomly generated within `radius_range`. For polygon obstacles, `avg_radius_range`, `irregularity_range`, `num_vertices_range`, and `spikeyness_range` defines which type of polygon will be generated. See [random_generate_polygon](#irsim.lib.algorithm.generation.random_generate_polygon) for details.
+
+
+The details of the parameters are listed in the [YAML Configuration](../yaml_config/configuration/)
 
-```python
+:::{tip}
+To generate a convex polygon, you can set `spikeyness_range` to [0, 0] or set `is_convex` to True.
+:::

doc/source/usage/configure_lidar2d.md

@@ -77,7 +77,7 @@ To configure the 2D LiDAR sensor, the sensor name of `lidar2d` should be defined
 - **number**: The number of beams.
 - **alpha**: The transparency of the laser beam.
 
-A full list of parameters can be found in the [YAML Configuration](#../get_started/configuration/).
+A full list of parameters can be found in the [YAML Configuration](#../yaml_config/configuration/).
 
 
 ## Advanced Configuration with noise

doc/source/usage/configure_robots_obstacles.md

@@ -186,6 +186,6 @@ The demonstration of the multiple robots and obstacles in the simulation are sho
 ```
 
 :::{note}
-- The `distribution` parameter specifies how the robots and obstacles are distributed within the environment. Options include `'manual'` and `'random'`. Details are provided in the [YAML Configuration](#../get_started/configuration)
+- The `distribution` parameter specifies how the robots and obstacles are distributed within the environment. Options include `'manual'` and `'random'`. Details are provided in the [YAML Configuration](#../yaml_config/configuration/)
 :::
 

doc/source/usage/make_environment.md

@@ -29,7 +29,7 @@ world:
 
 ## Explanation 
 
-The `world` section specifies the properties of the world. The `height` and `width` parameters specify the size of the world. The `step_time` parameter specifies the time step for the simulation. The `sample_time` parameter specifies the time step for rendering and data extraction. The `offset` parameter specifies the offset of the world on the x and y axes. The `control_mode` parameter specifies the control mode of the simulation. The `collision_mode` parameter specifies the collision mode of the simulation. The `obstacle_map` parameter specifies the path of the obstacle map. Details of the parameters can be found in the [YAML Configuration](#../get_started/configuration).
+The `world` section specifies the properties of the world. The `height` and `width` parameters specify the size of the world. The `step_time` parameter specifies the time step for the simulation. The `sample_time` parameter specifies the time step for rendering and data extraction. The `offset` parameter specifies the offset of the world on the x and y axes. The `control_mode` parameter specifies the control mode of the simulation. The `collision_mode` parameter specifies the collision mode of the simulation. The `obstacle_map` parameter specifies the path of the obstacle map. Details of the parameters can be found in the [YAML Configuration](#../yaml_config/configuration/).
 
 :::{tip}
 The default YAML configuration file is same as the name of python script. Thus, if you create a python script named `test.py`, the default YAML configuration file is `test.yaml`. And you can simply use `irsim.make()` to create the environment. Please place the YAML configuration file in the same directory as the python script.

doc/source/usage/save_animation.md

@@ -11,7 +11,7 @@ You can render the environment by calling the [env.render()](#irsim.env.env_base
 
 ## Save the animation
 
-You can save the animation of the simulation as a gif file very easily by setting the `save_ani` to be `True` in the `make()` function:
+You can save the animation of the simulation as a gif file very easily by setting the `save_ani` to `True` in the `make()` function:
 
 ```python
 
@@ -41,7 +41,7 @@ The principle of the animation generation is to save the images of each frame an
 
 ## 3D Plot
 
-You can simply set the `projection` parameter to be `3d` in `irsim.make` function to render the 3D plot of the simulation. The example is shown below:
+You can simply set the `projection` parameter to `3d` in `irsim.make` function to render the 3D plot of the simulation. The example is shown below:
 
 ```python
 

irsim/lib/algorithm/generation.py

@@ -6,7 +6,7 @@
 
 def random_generate_polygon(
     number=1,
-    center_range=[0, 0, 10, 10],
+    center_range=[0, 0, 0, 0],
     avg_radius_range=[0.1, 1],
     irregularity_range=[0, 1],
     spikeyness_range=[0, 1],

irsim/usage/12dynamic_obstacle/dynamic_obstacle.yaml

@@ -24,9 +24,9 @@ obstacle:
     shape: {name: 'circle', radius: 1.0}  # radius
     state: [5, 5, 0]  
 
-  - shape: {name: 'rectangle', length: 1.5, width: 1.2}  # radius
-    state: [6, 5, 1] 
+  # - shape: {name: 'rectangle', length: 1.5, width: 1.2}  # radius
+  #   state: [6, 5, 1] 
 
-  - shape: {name: 'linestring', vertices: [[5, 5], [4, 0], [1, 6]] }  # vertices
-    state: [0, 0, 0] 
-    unobstructed: True
+  # - shape: {name: 'linestring', vertices: [[5, 5], [4, 0], [1, 6]] }  # vertices
+  #   state: [0, 0, 0] 
+  #   unobstructed: True

irsim/util/util.py

@@ -515,7 +515,7 @@ def random_point_range(range_low=[0, 0, -pi], range_high=[10, 10, pi]):
     if isinstance(range_low, list):
         range_low = np.c_[range_low]
 
-    if isinstance(range_low, list):
+    if isinstance(range_high, list):
         range_high = np.c_[range_high] 
 
     return np.random.uniform(range_low, range_high)

irsim/world/object_base.py

@@ -328,7 +328,7 @@ def step(self, velocity=None, **kwargs):
         """
 
         if self.static or self.stop_flag or self.kf is None:
-            self._velocity = np.zeros_like(velocity)
+            self._velocity = np.zeros(self.vel_shape)
             return self.state
         else:
             self.pre_process()