12345qiupeng
/
gtsam
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
849 Commits
1 Branch
0 Tags
79 MiB
C++ 59.9%
Jupyter Notebook 23.8%
Python 9.1%
MATLAB 3%
CMake 2%
Other 2%
 
 
 
 
 
 
Go to file
Frank Dellaert 3247751b5d Major check-in: there are now two interchangeable implementations of VectorConfig. 
VectorMap uses a straightforward stl::map of Vectors. It has O(log n)
insert and access, and is fairly fast at both. However, it has high overhead
for arithmetic operations such as +, scale, axpy etc...

VectorBTree uses a functional BTree as a way to access SubVectors
in an ordinary Vector. Inserting is O(n) and much slower, but accessing,
is O(log n) and might be a bit slower than VectorMap. Arithmetic operations
are blindingly fast, however. The cost is it is not as KISS as VectorMap.

Access to vectors is now exclusively via operator[]
Vector access in VectorMap is via a Vector reference
Vector access in VectorBtree is via the SubVector type (see Vector.h)

Feb 16 2010: FD: I made VectorMap the default, because I decided to try
and speed up conjugate gradients by using Sparse FactorGraphs all the way.
 		2010-02-17 03:29:12 +00:00
CppUnitLite Removing dependency hack speeds up compilation 2010-01-20 20:47:15 +00:00
colamd Removing dependency hack speeds up compilation 2010-01-20 20:47:15 +00:00
cpp Major check-in: there are now two interchangeable implementations of VectorConfig. 2010-02-17 03:29:12 +00:00
doc Math doc now matches code but doc is fairly rough right now 2010-01-26 20:54:03 +00:00
m4 Fixed autotools files for GSL inclusion so that they don't include the wrong BLAS implementation when ATLAS is enabled 2010-01-31 18:26:18 +00:00
matlab case change: SharedGaussian and SharedDiagonal are now classes with their own header file. Needed for MATLAB TORO hail Mary 2010-01-22 17:36:57 +00:00
wrap case change: SharedGaussian and SharedDiagonal are now classes with their own header file. Needed for MATLAB TORO hail Mary 2010-01-22 17:36:57 +00:00
.cproject Major check-in: there are now two interchangeable implementations of VectorConfig. 2010-02-17 03:29:12 +00:00
.project Major check-in: there are now two interchangeable implementations of VectorConfig. 2010-02-17 03:29:12 +00:00
AUTHORS Fixing directory structure 2009-08-21 22:23:24 +00:00
COPYING Fixing directory structure 2009-08-21 22:23:24 +00:00
INSTALL add some documentation 2010-01-27 05:22:17 +00:00
LICENSE Fixing directory structure 2009-08-21 22:23:24 +00:00
Makefile.am copy libcolamd during installation 2010-02-01 00:11:28 +00:00
README Added USAGE file with a short tutorial 2009-11-13 14:01:51 +00:00
USAGE Added USAGE file with a short tutorial 2009-11-13 14:01:51 +00:00
autogen.sh Fixing directory structure 2009-08-21 22:23:24 +00:00
configure.ac Fixed issue with configure script for enabling gsl and atlas 2010-01-27 14:37:45 +00:00
myconfigure add gsl and atlas to configure 2010-01-27 05:15:52 +00:00

README 

README - Georgia Tech Smoothing and Mapping library
---------------------------------------------------

What is GTSAM ?

  GTSAM is a library of C++ classes that implement smoothing and
  mapping (SAM) in robotics and vision, using factor graphs and Bayes
  networks as the underlying computing paradigm rather than sparse
  matrices. A set of MATLAB mex wrappers is included.

GTSAM is not (yet) open source: See COPYING & LICENSE

Directory structure:

  cpp	  C++ source 
  matlab  MATLAB proxy classes and wrappers


Boost Dependencies:
------------------
The GTSAM library is based on the 'Boost C++ Libraries' which can be
found here: http://www.boost.org/. 
Donwload the lates version and extract Boost in any place. To this
place the compiler will be linking.

- On Linux BOOST can also be installed with a packaged manager.
- On Mac OS Mac Port can be used.

For example the Boost path could be '/opt/local/include/' on a typical
Mac system, where you should be able to find one of the header files:
/opt/local/include/boost/config.hpp 

If your boost files are on a different place change the path according
to your path.

The path to the Boost files can be set as an environmental variable in
the startup scrip. For a Bash shell the startup file is ~/.bashrc
put the following command in this file:
export BOOST_DIR=/opt/local/include/ 

Installation:
-------------
To finally install the library go into the directory where you unpacked the 
GTSAM library, run the command below for example:
$]./configure --with-toolbox=$HOME/toolbox/ --with-boost=/opt/local/include/

where the path after --with-toolbox should point to the directory you want to have the gtsam
matlab scripts installed in.

This command will configure the makefile for compiling the GTSAM library.

The 'toolbox' flag sets the path where you want to install the GTSAM Matlab Toolbox.
You have to set it to an existing directory. After successful installation there
will be a gtsam directory with all Matlab GTSAM Toolbox files.

The 'boost' flag sets the path where you installed or copied the BOOST C++ Library.
The path has to be set to the top boost directory. In this directory there are a bunch
of folders (e.g. boost, doc, libs ....).
Set the path to this folder.
  

After configure you makefile you have to compile the library
Type:
$] make
$] make install

Built-in Unit Tests:
----------------
There is one more optional step in which you can invoke the unit tests included in the gtsam libraries. 
$] make check
By verifying all the test results are positive, you can make sure that the functionalities of the gtsam
libraries are correct.

The toolbox directory flag is where you want to compile the GTSAM Matlab toolbox.

Compile Matlab Toolbox:
-----------------------
1) Start Matlab
2) Go to File->Set Path and add the toolbox directory where you installed the
   GTSAM Matlab Toolbox
3) Change your current directory to the GTSAM Matlab Toolbox
4) Type 'make_gtsam' at the Command Window

Run Matlab Unit Tests:
-----------------------
In the matlab command window, change directory to $gtsam/matlab and then type 'run_tests', which will 
invoke the matlab unit tests.