it should raise error (#594)

* it should raise error

* restore

* raise error?

* test pass merge

documentation/source/sensors.ipynb

@@ -70,7 +70,9 @@
    "execution_count": null,
    "id": "425d66f9-118f-4b91-a343-ef1385281ba8",
    "metadata": {
-    "tags": []
+    "tags": [
+     "skip_execution"
+    ]
    },
    "outputs": [],
    "source": [
@@ -85,7 +87,6 @@
     "               show_terrain=not os.getenv('TEST_DOC'),\n",
     "               sensors=dict(rgb=[RGBCamera, *size]))\n",
     "\n",
-    "\n",
     "env = BaseEnv(env_cfg)\n",
     "env.reset()\n",
     "print(\"Available sensors are:\", env.engine.sensors.keys())\n",
@@ -101,7 +102,9 @@
    "execution_count": null,
    "id": "1dd842e2-c89f-4715-af17-4a2d00f7bdd9",
    "metadata": {
-    "tags": []
+    "tags": [
+     "skip_execution"
+    ]
    },
    "outputs": [],
    "source": [

documentation/source/system_design.ipynb

@@ -72,7 +72,7 @@
    "id": "e8b4bd74-6874-4dcb-ba93-78dc0172189b",
    "metadata": {},
    "source": [
-    "As we use the bullet engine for underlying physics simulation, the `self.engine.step()` is actually a wrapper for `BulletWorld.doPhysics(dt, 1, dt)`. It will forward the physics world by a time period, which is set to `0.1s` by default. The result is the vehicle's speed will be increased to $5\\ m/s$ at state $s'$ if it has a $4.9\\ m/s$ speed with a $1m\\ /s^2$ acceleration before advancing the simulation. \n",
+    "As we use the bullet engine for underlying physics simulation, the `self.engine.step()` is actually a wrapper to repeatedly execute `BulletWorld.doPhysics(dt, 1, dt)` several times which is determined by `decision_repeat`. It thus will forward the physics world by a time period, which is `dt x decision_repeat = 0.02 x 5 = 0.1s` by default. The result is the vehicle's speed will be increased to $5\\ m/s$ at state $s'$ if it has a $4.9\\ m/s$ speed with a $1m\\ /s^2$ acceleration before advancing the simulation.\n",
     "\n",
     "Before advancing the physics world, there may be some preparations required. For example, if the external input contains signals for controlling a certain vehicle, the steering angles and the throttle value should be set on the target vehicle before stepping it. Thus all these preprocessings are finished by calling `self.engine.before_step()`. \n",
     "\n",
@@ -81,9 +81,8 @@
     "If we dive deeper to check the `engine.before_step()`, `engine.step()`, and `engine.after_step()`. We will find that all three functions share the same structure. Take the `before_step()` function as the example. What the `engine.before_step()` does is actually sequentially executing `mgr.before_step()` for each manager.  \n",
     "```python\n",
     "def before_step(self):\n",
-    "    for _ in range(config[\"decision_repeat\"]):\n",
-    "        for mgr in managers.values():\n",
-    "            mgr.before_ste() # or mgr.step() or mgr.after_step()\n",
+    "    for mgr in managers.values():\n",
+    "        mgr.before_step() # or mgr.step() or mgr.after_step()\n",
     "```\n",
     "Thus the `engine.step()` is to sequentially execute `before_step()`, `step()`, and `after_step()` for all managers.\n",
     "In addition, a simulation needs an initial state as the start point of everything. \n",