go-Corona-go

The project aims to aid safe travelling for people by providing the safest path from a source to a destination.

Currently the implementation only consists of the locality in Delhi, but it can further implemented such that it takes the locality of the user as a parameter.

DEMO VIDEO

https://youtu.be/lcXM4QS5a_Y

graph-parser-osm

Consists of Java program that parses OSM (OpenStreetMap) XML files into json graph representation.

• It was used to detect all the roads present in the given area (in the form of map.osm).
• Input given is the .osm file and output is the json file consisting of nodes which are intersection of roads with latitudes, longitudes, unique index and weights.
• This can be treated as a black-box or a tool to get the nodes of the graph in the locality.

Instructions

• Follow the steps:

1. Both in parent as well as in covi-secure directory

npm install 

2. Open two terminal windows:

• In one window stay in the parent directory & run:

nodemon

In other window, open covi-secure directory & run:

npm start

Implementation

• index.js is the master file.
• Given the starting & ending latitude and longitude we find the nearest node to these locations.
• The project currently consists of fake crowd data / simulation. It can easily be used on real data.
• infected.json consists of the latitude and longitude of the infected people which has been generated by fake_data.py.
• The real world data can also be collected and refreshed in time intervals to give real time threat factor for the paths.
• Applying the modified Djikstra algorithm by adding risk factor in the weights (distance b/w nodes), gives us the shortest and safest path.

    var newGraph = [];
    lambda = 0.1 // Change this accordingly
    for (var i = 0; i < Object.keys(g).length; i++) {
        var adjlist = [];
        var child = g[i].e;
        for (var j = 0; j < Object.keys(child).length; j++) {
            adjlist.push({
                'i': child[j].i,
                'w': lambda * child[j].w + riskValue[j] + riskValue[i] // Risk factor addded in the weights
            })
        }
        newGraph.push(adjlist);
    }

• The distance between two points are not taken directly by the euclidean distance formula but the actual formula between two coordinates is applied for far more precision.

function getDistanceFromLatLonInKm(lat1, lon1, lat2, lon2) {
    var R = 6371; // Radius of the earth in km
    var dLat = deg2rad(lat2 - lat1); // deg2rad below
    var dLon = deg2rad(lon2 - lon1);
    var a =
        Math.sin(dLat / 2) * Math.sin(dLat / 2) +
        Math.cos(deg2rad(lat1)) * Math.cos(deg2rad(lat2)) *
        Math.sin(dLon / 2) * Math.sin(dLon / 2);
    var c = 2 * Math.atan2(Math.sqrt(a), Math.sqrt(1 - a));
    var d = R * c; // Distance in km
    return d;
}

Locality for which it is tested