Update pereption README.md and add FAQ document (#341)

* Update pereption README.md and add FAQ document

* Update document

* Fix error in document

docs/howto/how_to_run_perception_module_on_your_local_computer.md

+# How to Run Perception Module on Your Local Computer
+
+The perception module requires Nvidia GPU and CUDA installed to run the CNN Segmentation algorithm with Caffe. We have already installed the CUDA and Caffe libraries in the released docker. However, the pre-installed Nvidia GPU driver in the released docker image may be not compatible to your host machine for offline debugging and simulation, which will make the percetion module fail to run. We suggest to reinstall the exactly same version of Nvidia driver in the docker image as the one installed in the host machine, and build Apollo with GPU option.
+
+We provide a step-by-step instruction on running perception module with Nvidia GPU as below:
+
+1. Modify the script `./docker/scripts/dev_start.sh` to mount the library and source code of linux kernel from the host machine, by adding two option lines in command `docker run -it`:
+```
+-v /usr/src:/usr/src 
+-v /lib/modules:/lib/modules
+```
+
+2. Start the released docker image and get into docker with root authority: 
+``` 
+./docker/scripts/dev_start.sh
+docker exec -it apollo_dev /bin/bash
+```
+
+3. Uninstall the pre-installed Nvidia driver in docker:
+```
+sudo apt-get autoremove --purge nvidia-*
+```
+
+4. Download official Nvidia driver installation file which should be the exactly same version as the one installed in the host machine. We recommand the version of 375.39:
+```
+wget http://us.download.nvidia.com/XFree86/Linux-x86_64/375.39/NVIDIA-Linux-x86_64-375.39.run
+```
+
+5. Install Nvidia driver in docker (using Linux-x86-375.39 version as example):
+```
+chmod +x ./NVIDIA-Linux-x86_64-375.39.run
+./NVIDIA-Linux-x86_64-375.39.run --no-opengl-files -a -s
+```
+
+6. Commit a new docker image (in host):
+```
+docker commit CONTAINER_ID apolloauto/apollo:NEW_DOCKER_IMAGE_TAG
+```
+
+7. Start the new docker image (in host) and get into docker:
+```
+./docker/scripts/dev_start.sh NEW_DOCKER_IMAGE_TAG
+./docker/scripts/dev_into.sh
+```
+
+8. Build Apollo with GPU option (in docker):
+```
+./apollo.sh build_gpu
+```
+or 
+```
+./apollo.sh build_opt_gpu
+```
+
+Now the perception module can be running in GPU mode with command `./scripts/perception start`. Please note that the Nvidia driver should be installed appropriately as shown above even if the perception moduled is running in Caffe CPU_ONLY mode (i.e., using `./apollo.sh build` or `./apollo.sh build_opt` to build the perception module). Please note that the Nvidia driver should be installed appropriately as shown above even if the perception moduled is running in Caffe CPU_ONLY mode (i.e., using `./apollo.sh build` or `./apollo.sh build_opt` to build the perception module).
\ No newline at end of file

docs/specs/3d_obstacle_perception.md

@@ -263,7 +263,7 @@ explains the parameter usage and default values for CNN Segmentation.
  |height_thresh                      |If it is non-negative, the points that are higher than the predicted object height by height_thresh will be filtered out in the post-processing step.	|0.5 meters |
  |min_pts_num                        |In the post-processing step, the candidate clusters with less than min_pts_num points are removed. |3   |
  |use_full_cloud                     |If it is set by true, all the points of the original point cloud will be used for extracting channel features. Otherwise only the points of input point cloud (i.e., the points after HDMap ROI filter) are used.   |true	|
- |grid_id                            |The ID of the GPU device used in the CNN-based obstacle prediction step.			  |0	      |
+ |gpu_id                             |The ID of the GPU device used in the CNN-based obstacle prediction step.			  |0	      |
  |feature_param {width}              |The number of cells in X (column) axis of the 2D grid.					  |512	      |
  |feature_param {height}             |The number of cells in Y (row) axis of the 2D grid. 					  |512        |
  |feature_param {range}              |The range of the 2D grid with respect to the origin (the LiDAR sensor).			  |60 meters  |

modules/perception/README.md

@@ -13,4 +13,6 @@
   * 3D obstacle tracks with heading and velocity information (ROS topic _/apollo/perception/obstacles_)
 
 ## Instruction
-  Before running the 3D obstacle perception program, please select the appropriate HD map by setting the option `--map_dir` in the global configuration file `modules/common/data/global_flagfile.txt`. After that, you may setup the general settings in the configuration file `modules/perception/conf/perception.conf`. Then, you can launch the perception program by using the command `./scripts/perception start` or enabling the perception button in HMI. The command of stopping perception is `./scripts/perception stop`. In addition we provide some demo data for developers. Please download the demo data from our [Open Data Platform](https://console.bce.baidu.com/apollo/task/download).
\ No newline at end of file
+  Before running the 3D obstacle perception program, please select the appropriate HD map by setting the option `--map_dir` in the global configuration file `modules/common/data/global_flagfile.txt`. After that, you may setup the general settings in the configuration file `modules/perception/conf/perception.conf`. Then, you can launch the perception program by using the command `./scripts/perception start` or enabling the perception button in HMI. The command of stopping perception is `./scripts/perception stop`. In addition we provide some demo data for developers. Please download the demo data from our [Open Data Platform](https://console.bce.baidu.com/apollo/task/download).
+
+  **Note**: It requires Nvidia GPU and CUDA installed to run the perception module with Caffe. We have already installed the CUDA and Caffe libraries in the released docker. However, the pre-installed Nvidia GPU driver in the released docker image may be not compatible to your host machine for offline debugging and simulation, which will make the percetion module fail to run. We suggest to reinstall the exactly same version of Nvidia driver in the docker image as the one installed in the host machine, and build Apollo with GPU option (i.e., using `./apollo.sh build_gpu` or `./apollo.sh build_opt_gpu`). Please see the detailed instruction in [How to Run Perception Module on Your Local Computer](https://github.com/ApolloAuto/apollo/blob/master/docs/howto/how_to_run_perception_module_on_your_local_computer.md). 
\ No newline at end of file