jackieh / path-estimation

A C++ library for two-dimensional path smoothing. Given a set of coordinate pairs with timestamps that approximate a path of movement, this library provides functionality to reduce noise in the data and guess the true path. Includes a test suite and test generation.

Path Estimation

This repository contains code for a C++ library which predicts a path for two-dimensional movement, given a set of data points which contain a coordinate pair and a timestamp.

Dependencies

The library itself has the following dependencies.

• Armadillo for matrix operations

The test code which uses the library has the following dependencies.

• PLplot for path visualization
• JsonCpp for JSON parsing

Building and testing

The Makefile has a test target to compile a test program that uses the library. The run target will run the test code on an example input.

For more thorough testing, python test/driver.py --all can be run to produce results for every available test case.

Documentation

The docs target in the Makefile will generate doxygen documentation.