Cellular phones have made significant strides in recent years and are high-performance multi-functional mobile devices. Although pedestrians with cell phones are have access to Global Positioning System (GPS) services, a number of challenges must be solved for fully functional pedestrian navigation systems. These include GPS errors in the location measurement, methods to generate simplified maps by removing unnecessary information from the display, and GPS location measurements in places other than roads (e.g., indoors or a wide open space such as a city plaza). The Map Group is working on these issues from diverse perspectives.
A map-based navigation system is useful, especially when visiting a place for the first time. However, if a user cannot correctly read the information on the map, he or she may become lost and unable to reach the destination. The Map Group engages in research to develop a clear pedestrian navigation system by considering what information is necessary to reach the destination and how to present such information so that the user can recognize the information. For example, we have developed maps where a landmark is highlighted (colored and circled) at the corner where a turn is necessary, allowing the route to be easily identified. We are implementing such an understandable navigation system to smart phones and other devices with the goal of commercialization.
GPS Positioning in some places has an error of 25 – 50 m, and the error may be as high as 150 m in the worst cases due to an insufficient number of positioning satellites. The figure below shows examples where radio waves do not reach and multi-paths (a condition where radio waves are reflected on buildings and ground to interfere with accurate positioning) are present. To resolve these issues, an error correction method is necessary. We have proposed a three-step correction method. First, a cell phone camera is used to take a picture of a road sign. Second, the sign is identified. Finally, the sign is compared to a correlation database of maps and road signs. Currently, we are developing a project for commercialization.
A major issue in indoor navigation is the inability to use GPS. Another issue is that a modeled map is required to search a route between two points. For an outdoor space, a graph model can be used with a network structure of roads, but indoor spaces have a special structure without roads. On the other hand, a visibility graph can be used to map indoor spaces. In a visibility graph, a graph model places a node on each apex of an obstacle, and branches from these nodes that do not go across obstacles are used as links. We are developing an indoor navigation system using a visibility graph.
GPS is generally used for positioning outdoors because its accuracy significantly decreases indoors due to signal loss. A positioning method based on the distance from an access point to a LAN has been proposed, but it requires the position of the access point to be acquired in advance. Pedestrian Dead Reckoning (PDR) is a positioning method that does not rely on other facilities. Instead, PDR uses acceleration sensors, geomagnetism sensors, and gyro sensors to measure the displacement from a reference point to measure its own position. Recent advancements in sensor technologies have realized small and multiple sensors, which can be found in many mobile devices, enabling the position of a person (holding a device) to be determined indoors. We are developing on a PDR positioning method to realize a navigation system for indoor use.
An issue with displaying a map on a cell phone is that the phone’s small screen makes it difficult to read. One solution is to generate a simpler map, which is an abridged version of a regular map where extraneous information has been removed. Simplifying a map makes it easier to read necessary geographical information. Many simplified maps are manually generated by designers. We are investigating an automatic simplified map generation methodology with the goal of automatically generating easily recognizable simplified maps.
Togawa Laboratory