Add make config file for Linux system, update readme.

Make-linux

+# set directory of external packages
+INCDIR = /usr/include
+LIBDIR = /usr/lib
+MATDIR = /usr/local/MATLAB/R2016b
+
+
+# set all compiler and linker flags
+OPT = -O3
+#EXTRA_OPT = -fwhole-program
+
+# for armadillo
+ARMA = -DARMA_NO_DEBUG -DARMA_DONT_USE_WRAPPER
+
+CXX = g++
+
+INCS = -I$(INCDIR)/boost \
+	-I$(INCDIR)/armadillo_bits \
+	-I$(MATDIR)/extern/include \
+
+LIBS = -lblas -llapack -lmat -leng -lmx
+LIBTEST = -lboost_unit_test_framework
+
+CXXFLAGS = -g \
+	$(ARMA) $(OPT) \
+	-std=c++11
+
+LDFLAGS =-L$(MATDIR)/bin/glnxa64
+	
+

README.md

-# RObustly Complete control Synthesis (ROCS)
+# RObustly Complete control Synthesis (ROCS) #
 
 **ROCS** is a robust symbolic control synthesis tool for nonlinear systems. It provides two control synthesis methods from linear temporal logic (LTL) formulas:
 - Abstraction-based control synthesis (to be added): state and control space are uniformly discretized.
 - Specification-guided control synthesis: state space are partitioned adaptively with respect to given specifications and local dynamics of the systems.
 
-## Installation
-### Prerequisite packages and softwares
+**ROCS** is currently a package of c++ APIs, which only need to be included and no recompiliation required.
+
+
+
+## Installation ##
+### Prerequisite packages and softwares ###
  - boost c++ libraries (http://www.boost.org): libraries for graph searching, ode solvers;
  - armadillo (http://arma.sourceforge.net): a c++ linear algebra library;
  - matlab: control simulation and graphics.
-### Configuration
- - Update the environment variable DYLD_LIBRARY_PATH in terminal:
+
+### Configuration of Matlab ###
+ - Matlab 2017a and earlier versions are supported.
+ - Follow the instruction Matlab documentation on **Set run-time library path on Mac and Linux systems**, e.g.,
+   - On Mac OSX: update the environment variable DYLD_LIBRARY_PATH by placing the following commands in a startup script (~/.profile):
+```
+export DYLD_LIBRARY_PATH=“matlabroot/bin/maci64:matlabroot/sys/os/maci64:$DYLD_LIBRARY_PATH”
 ```
-export DYLD_LIBRARY_PATH=“/Applications/MATLAB_R2016a.app/bin/maci64:/Applications/MATLAB_R2016a.app/sys/os/maci64:$DYLD_LIBRARY_PATH”
+   - On Linux:
 ```
+export LD_LIBRARY_PATH=“matlabroot/bin/glnxa64:matlabroot/sys/os/glnxa64:$LD_LIBRARY_PATH”
+```
+
+### Installation of **Armadillo** ###
+The installation differs on Mac OSX and Linux.
+ - For Mac users, recommend to use *Homebrew* with simple command in terminal:
+ ```
+ $brew install armadillo
+ ```
+ - For Linux users, packages such as *cmake*, *openBLAS*, *LAPACK* and *ARPACK*. Follow the instruction on http://arma.sourceforge.net/download.html.
+
 
 
-## Usage
-- Clone this repositary. Source code will be downloaded and ready for use.
+## Usage ##
+- Clone this repositary.
 - Make sure the prerequisites are installed.
-- Go to the folder *examples*, modify the *Makefile* according to your actual installed paths of prerequisites, e.g.
-```
-BOOSTROOT = /usr/local/Cellar/boost/1.62.0
-MATLABROOT = /Applications/MATLAB_R2016a.app
-LINALGROOT = /usr/local/Cellar/armadillo/7.500.2
-```
-- Type `make` in command window to compile examples. For example, for dcdc example, type `make dcdc` in command window.
-- Execute compiled examples by typing the name of the executable, e.g. `./dcdc`.
+- Prepare the configuration file for build:
+  + The default configuration file *configMake* in the root path is for Mac OSX, the user only need to modify the paths for the external packages, e.g.,
+  ```
+  BSTDIR := /usr/local/opt/boost
+  MATDIR := /Applications/MATLAB_R2016a.app
+  LADIR := /usr/local/opt/armadillo
+  ```
+  + for Linux users, rename the file *Make-linux* to *configMake*. If all the prerequisites are installed in the default paths, only modify the matlab path, e.g., 
+  ```
+  MATDIR = /usr/local/MATLAB/R2016b
+  ```
+- To run examples, go to the folder *examples*:
+  + To build all the examples: type `$ make examples` in terminal.
+  + To build one of the examples: go to the corresponding subfolder and type `$ make` in terminal.
+- Execute compiled examples by typing the name of the generated executable, e.g. `./dcdc`. **Data will be generated in each subfolder corresponding to the specific example**.
 - Run *simulation.m* in the same example folder using matlab.
 
-## Examples
+
+
+## Examples ##
 Simulations of the examples can be found under the folder *examples*:
 - *examples/dcdc*: DCDC converter invariance control.
 - *examples/car*: Robot car reach avoid control: reach a target region while avoiding obstacles.
 - *examples/ipdl*: Inverted pendulum reach stay control: reach and stay in the upright position.
-- *examples/poly*: Polynomial systems region of attraction computation.
-- *examples/temperature*: Room temperature control (4-mode system): reach and stay in desired room and heater temperatures.
+- *examples/temp*: Room temperature control (4-mode system): reach and stay in desired room and heater temperatures.
+
 
-## Create your own examples
+
+## Create your own examples ##
 To create your own examples, you have to provide a *main file* specifying all necessary problem settings and an additional file of system dynamics. For the DCDC converter example,
 - main file: *DemoDcdc.cpp*
 - vectorfield file: *dcdc.hpp*
 
-## Features included
+
+
+## Features included ##
 - Basic invariance and reachability control for discrete-time systems.
 - Buchi, coBuchi and reach-stay (a simpler version of coBuchi) control synthesis algorithms.
 
-## Features under developement
+
+
+## Features under developement ##
 - Control synthesis from more complex LTL formulas.
 - Integrate continuous-time validated ode solver.
 
 
-## Note
-- Only tested on Mac OSX 10.11/12.
\ No newline at end of file
+
+## Note ##
+Tested on
+- Mac OSX 10.11/12.
+- Ubuntu 16.04.3 LTS.
\ No newline at end of file