Use of AI in commuting
- 1. “USE OF AI IN COMMUTING” Prepared By: Bansi Mehta (15IT050) Department of Information Technology Chandubhai S. Patel Institute of Technology IT-321 | Sem-6 | Department of Information Technology | CSPIT
- 2. OUTLINE • Importance. • Adding data to database. • Storing data. • Practical approach. • Working. • Route determination algorithm. • Issues. IT-321 | Sem-6 | Department of Information Technology | CSPIT
- 3. IMPORTANCE OF AI IN COMMUTING • Statics: Rush-hour traffic delays a commuter by 42 hrs. on an average in a year, which is more than a full work week, with an estimated $160 billion in lost productivity. • Clearly, there’s massive opportunity here for AI to create a visible impact in every person’s life. • Solution: Google Maps. IT-321 | Sem-6 | Department of Information Technology | CSPIT
- 4. ADDING DATA TO DATABASE • For street view: • A digital video camera with 11 lenses mounted to a geodesic globe that sits on top of Google's fleet of photo vehicles. • The camera simultaneously captures high resolution images in every direction to create a 360-degree view of its surroundings. • It’s built-in GPS processor records geographical coordinates and matches them to the video. • The digital video and GPS files are then stored on hard drives for processing. IT-321 | Sem-6 | Department of Information Technology | CSPIT
- 5. CONTD. • To convert motion video footage to usable still images, a vertical slice of the changing view directly in front of each camera must be regularly cropped from the moving image plane. • When these vertical slices, from multiple perspectives, are stitched together in a linear sequence, a panoramic Street View is created. • Street names, locations on the street are attached with it and considering the privacy the license plate of vehicles and faces on individuals are blured. • User can add data to Google maps which are analyzed for its authencity and then added to Google maps. IT-321 | Sem-6 | Department of Information Technology | CSPIT
- 6. STORING DATA • GIS- a special kind of geological database. • It stores data in the form of longitude, latitude and elevation of specific points. • A road is a sequence of two or more of these points. • Straight road: 2 points. • Curvy road: Multiple points. • Inside GIS it's really just a sequence of straight lines or polyline. IT-321 | Sem-6 | Department of Information Technology | CSPIT
- 7. CONTD. • Some data will be attached to each of these lines or polylines, such as the name of the road, the speed limit, whether it's a one-way street, and perhaps traffic information. • So, basically a road is stored as a vector map and this is how a road is identified in Google maps. • Vector maps means every feature is either point, line, or polygon. It means all roads are lines, and landmarks are points. IT-321 | Sem-6 | Department of Information Technology | CSPIT
- 8. PRACTICAL APPROACH • While finding shortest path it is not practical to consider every nook and corner of the city. • A that a hierarchical superstructure can be used. • For example, along with splitting a highway into one node/edge per exit they may also have 'longer' edges that connect exits at major cities, or edges that connect major traffic hubs within cities to the highways close to that city in one step. • The algorithm that we follow should first decides on which level of the hierarchy it wants to operate for different parts of the route. • In that way, the algorithm can explore longer paths without having to explore every back alley of each city. IT-321 | Sem-6 | Department of Information Technology | CSPIT
- 9. HOW DOES GOOGLE MAPS WORK? • Algorithms of computational geometry to organize the map data and retrieve it efficiently. • Algorithms to draw maps. • project latitude and longitude coordinates, fill the polygons, place names for streets, cities, businesses, parks etc. • Algorithms to understand queries from users. • Algorithms to process GPS signals. • Algorithms to perform geocoding, converting addresses to points and reverse geocoding. IT-321 | Sem-6 | Department of Information Technology | CSPIT
- 10. ROUTE DETERMNING ALGORITHM • First obvious choice- Dijkstra’s algorithm. • Although it is the most promising algorithm, it won’t work here because of the complexity of the graph. Also the factors like cost, time, traffic density needs to be considered along with distance. • Bi-Directional- It means that the route is computed both forward from the origin and backward from the destination. • But the algorithm must determine when the two searches have met at a point on an optimum or near optimum route. IT-321 | Sem-6 | Department of Information Technology | CSPIT
- 11. CONTD. • A*- It is formulated in terms of weighted graphs. In case of Google map this weight is travel time. • Starting from a specific node(source node) of a graph, it constructs a tree of paths, expanding paths one step at a time, until one of its paths ends at the predetermined destination node. • At each iteration, A* needs to determine which of its partial paths to expand into one or more longer paths. • It does so based on an estimate of the cost to go to the goal node. • This function may vary according to the needs. IT-321 | Sem-6 | Department of Information Technology | CSPIT
- 12. ISSUES: • Other issues affecting the algorithms: • Detecting when the driver has gone off the route and needs a new route. • Special requirements depending on the type of transportation • Differences between states and countries that affect • Rules for using roads. • Language and addressing systems supported by geocoding. • 3D map data – organizing and IT-321 | Sem-6 | Department of Information Technology | CSPIT
- 13. THANK YOU