Intelligent Nautical Route Planning and Calculation by monkey6


More Info
									Intelligent Nautical Route Planning and Calculation
ME Chamberlain 20 February 2007


Introduction and Problem Statement

For hundreds of years shipping (in all senses, but especially the naval sense) has been the back bone of the global economy. Tons of cargo can be transported over vast distances. Like in any business, money changes hands during the shipping process. The shipping companies need a model to determine how much to charge customers for the shipping service they provide. This model would certainly need to take into account the distance that the ship (or ships) had to travel, as well as the time that the ships spent at sea. The focus of this project is in determining an optimal route (under certain conditions), the distance and the time it would take for a ship to travel between two points on the ocean. At first this problem might sound trivial, but there are a lot of factors to take into consideration. For instance, the route calculated must be optimal. An optimal route would be one that, to the largest extent possible, avoids certain dangerous or impassable areas. It would also need to be the shortest route. This might sound counter intuitive, if the shipping company charges per kilometer (or nautical mile) that the ship travels, a longer route would mean more money for them. There is a problem in that argument - time. The time it takes for a package to travel from one point on the globe to another directly affects how much the customer would be willing to pay, if a competitor could deliver it sooner. The longer it takes for a package to reach its destination, the less the customer is willing to pay for the service. The biggest challenge behind this system is the large amount of data. A world map is required. For the purposes of navigation, maps of a high level of detail and accuracy are required. These maps need to be loaded into primary memory (at least partially) to run an algorithm on the data. Finally execution time is an important factor. The system cannot take days to calculate routes, decisions need to made in a matter of hours.


Brief Overview of the Project Requirements

The program should be able to find an optimal route between two points. These points cannot simply be ports - it can be any two points on the ocean. The business situation is ever changing, and that could lead to goods being sold in the midst of a journey. That would require a ship to possibly change its destination in the middle of its trip.


The optimality of a route is determined by minimizing the distance between the two points whilst taking the following situations into account: 1. Land masses: A ship cannot pass through land, so any land masses need to be avoided, this includes the most obvious of the land masses, the continents. It also includes islands. 2. Shallow Waters: Certain ships, like oil tankers, cannot sail in shallow waters, especially if they are not carrying any cargo. Depending on the ship requirements, the algorithm would need to exclude certain shallow regions of the ocean. Reefs, like the The Great Barrier Reef are also impassable for certain vessels. 3. Narrow Regions: Channels, like the English Channel have narrow regions that not all ships can pass through. Based on the ship’s requirements the algorithm would need to exclude these areas when finding a possible route. 4. Weather: Storms, even if it does not sink a ship, could cause huge losses of cargo and damage to a ship. Certain regions of the ocean are marked as dangerous in some seasons. 5. Piracy: Some regions of the East African coast have become piracy hot spots. With some of the weapons available today vessels can easily be damaged and boarded. The program must have the functionality to define new problem regions, change existing problem regions or remove them. The program must be able to calculate routes in a reasonable amount of time. An upper bound would be half a day. After route calculation, some form of output representing the route must be given, possibly vector data.


The Challenges

This project was requested by a company. They have existing software, but it does not meet all of the requirements. This project could be implemented from scratch, or it can use the company’s existing software and add the optimal route calculation functionality. The challenges facing the project include: 1. Data structure: A data structure that is able to store large amounts of data, and is efficient for traversal, must be implemented. 2. Efficient Algorithms: Efficient shortest path or optimization algorithms need to be tested and compared against one another to determine the best one for this case. 3. Data Format for Danger Regions: A data format to store danger regions, like piracy hot spots, needs to be defined. A method for super imposing that information onto the world map needs to be defined. 4. Store Calculated Routes: Calculated routes could be stored for future use. 5. A Caching System: A caching system between the map data on secondary memory and primary memory could be implemented, depending on the performance of the operating system’s caching mechanism. Customized caching may lead to better performance or it could counteract the operating system’s caching. It would need to be investigated. 2

6. ”Real-Time”: The algorithm’s time requirements need not be completely real time, an upper bound of half a day is given for computation time. Another consideration would be a user interface. An existing 2D one could be used, or a new 2D or 3D interface can be implemented. The resource costs of a 3D system should be taken into account, as it could use valuable resources that could be better used by the route calculation algorithm. Other possible functionality that could be included if time permits it, is real time weather updates, or at least updates at set intervals. Global Positioning System (GPS) tracking support for the vessels could also be considered.


To top