Skip to content
Fast 3D Feature Map Updates for Automobiles
C++ Cuda CMake Shell
Branch: master
Clone or download
Latest commit ecefb98 Dec 22, 2019
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
CarMap_Cloud_Service Update README.md Oct 7, 2019
CarMap_Vehicle_Client
.gitignore first commit Sep 19, 2019
CarMap_Gif.gif Add files via upload Oct 8, 2019
README.md updated readme Dec 22, 2019
build_carmap.sh

README.md

CarMap-Fast 3D Feature Map Updates for Automobiles

Features:

  1. CarMap-generated 3D lean map (20x smaller map than ORB-SLAM2)
  2. CarMap's dynamic object filter to remove dynamic objects
  3. CarMap's semantically annotated 3D map
  4. CarMap's position-based robust feature matching
  5. CarMap's efficient map update operation
  6. CarMap's robust map segment stitching for unmapped regions
  7. CarMap-generated 3D map upload/download to the CarMap cloud service
  8. Feature map stitching for CarMap generated map

Build Instructions

There are two directories in the repo:

The vehicle client: CarMap_Vehicle_Client

The cloud service: CarMap_Cloud_Service

Both of them have similar dependencies and build instructions

Dependencies:

  1. Run the build_carmap.sh script to build the dependencies needed
sh build_carmap.sh
  1. Building OpenCV:
git clone https://github.com/opencv/opencv
cd opencv
git checkout 3.4.0
mkdir build && cd build
cmake -D CMAKE_BUILD_TYPE=Release -D CMAKE_INSTALL_PREFIX=/usr/local ..
make -j7
sudo make install
  1. Eigen 3.1.0 or higher

  2. Pangolin

git clone https://github.com/stevenlovegrove/Pangolin.git
mkdir build && cd build
cmake ..
make -j
  1. PCL 1.9.0

Configuring CarMap:

  1. First build CarMap_Cloud_Service
cd CarMap_Cloud_Service
mkdir build && cd build
cmake ..
make -j
  1. Build CarMap_Vehicle_Client
cd CarMap_Vehicle_Client
mkdir build && cd build
cmake ..
make -j
  1. Download sample dataset for CarMap_Vehicle_Client:
mkdir CarMap_Dataset && cd CarMap_Dataset

Navigate to the link (https://drive.google.com/uc?export=download&id=1eJ4yZGrQXC50Hp4Idie2I3K5XlyR-Eeb) and download the dataset

tar -xz CarMap_Dataset.tar.xz
cd ..

Running CarMap:

  1. Run CarMap Cloud Service
cd CarMap_Cloud_Service/build/
./server_stitch
cd ../../
  1. Run CarMap_Vehicle_Client
cd CarMap_Vehicle_Client/build/
./main

Instructions for Running CarMap

Both the vehicle client and the cloud service have a Config_File.yaml that dictates the operation mode of CarMap.

CarMap Cloud Service:

In this section, we describe what some of the fields in Config_File.yaml of CarMap Cloud Service are used for:

  1. All the ServerStitch.Path* variables are used to feed input and output paths to CarMap
  2. OneShotStitching: true for CarMap stitching and false for progressive relocalization
  3. ReceiveMaps: true for receiving map segments from a vehicle client and false to stitch two maps from disk
  4. CloudServiceIDAddress: this is the ip address of the cloud service (always starts with ip.)
  5. CloudServicePort: port address at the cloud service
  6. IntegrateDiff: true for map update using diff and false for map update via stitch

CarMap Vehicle Client:

In this section, we describe what some of the fields in Config_File.yaml of CarMap Vehicle Client are used for:

  1. OperationMode.OpCode:

    • 0 for a first time mapping session with no pre-loaded map
    • 1 for updating a pre-loaded map
    • 2 for using a pre-loaded map without updating it
  2. SaveMap/UpdateMap/LoadMap.StartFrom and StopAt: starting and ending frame numbers of input dataset

  3. Path variables are for input and output paths

  4. Constants:

    • RunLocalization: false for SLAM and true for running only localization (we use false by default)
    • SaveTrajectory: true for saving trajectory at output path
    • IsStereo: true for stereo camera frames as input
    • OrbVisualization: true to visualize the whole CarMap mapping process
  5. Mode:

    • RemoveDynamicObjects: true for using dynamic object filter
    • LoadDepth: true to load depth images from input_dataset/depth/ but false by default because we estimate depth from stereo matching
    • EvaluateSegmentation: true to evaluate segmentation accuracy using ground truth segmented images
    • RobustFeatureSearch: true to enable robust feature matching
  6. Online Mode Settings:

    • UploadToCloud: true to upload map segments/diffs to cloud service
    • UploadOnlyDiff: true to upload diff and false to upload whole map segments
    • CloudIPAddress and CloudPort: address of carmap cloud service
    • UploadFrameInterval: number of frames after which to upload map diff/segments to cloud
You can’t perform that action at this time.