Issue link: http://uwashington.uberflip.com/i/742071
Research Highlights The Federal Highway Administration (FHWA) requires states to inspect and inventory federal-aid highway system bridges every two years to assess bridge safety. However bridge inspection is inherently dangerous. Inspectors stand in platform trucks, bucket trucks, or under-bridge inspection vehicles in order to access bridge elements. Some inspections even require climbing, use of temporary scaffolding and ladders, or rescue boats should the inspector fall in the water. In addition to the danger to the inspector and vehicle operator, road users also face danger as traffic lanes on the bridge are closed or reduced during an inspection. The purpose of this research is to evaluate the opportunities of small Unmanned Aircraft Systems (UAS) technology to perform remote, visual bridge inspections. UAS are capable of flying a safe, pre-loaded flight path, and can carry high resolution digital cameras and other sensors. During flights, operators can view live video from the camera. In addition, overlapping digital photographs taken from the UAS can be mosaicked, georeferenced, and converted into 3D point clouds for detailed spatial inventorying. COST-EFFECTIVE BRIDGE SAFETY INSPECTIONS USING UNMANNED AERIAL VEHICLES (UAVS) The team of researchers on this project investigated available UAS, developed protocols for safe and efficient flights that meet Federal Aviation Administration rules, and conducted test flights of a bridge over the Willamette River in Oregon. Since then, the team was awarded a two-year research grant from the Oregon Department of Transportation to further the research and do additional test inspections of bridges that vary in size and material. The results of the flights conducted to date show that digital imagery collected from the UAS facilitate evaluation of the condition of the connections and joints on the bridge, identification of concrete spalling and rust, and mapping of the upstream and downstream banks beneath the bridge. Contact: Dan Gillins, dan.gillins@oregonstate.edu; Chris Parrish, chris.parrish@oregonstate.edu 2013 NHTSA data showed almost 5,000 pedestrian fatalities, 19% of which were pedestrians 65 and older, an age group that is overrepresented in suffering from vision or mobility impairments. This current project aims to evaluate the use of ultra-wideband radio signals for tracking a pedestrian as they cross an intersection. Such a capability would allow the traffic controller to provide navigational feedback to the pedestrian in case they veer out of the crosswalk, as well as alert approaching vehicles using Vehicle to Infrastructure technology. Several transceiver were purchased. This technology recently achieved 23cm accuracy in indoor-positioning and similar results are expected in outdoor environments. To test this hypothesis, we purchased an evaluation kit consisting of four transceivers mounted on printed circuit boards with support circuitry. Experiments were conducted with three units serving as "anchors" and the fourth as a "tag", carried by a "pedestrian". The figure below shows the results from placing anchors at each vertex of a triangle and separated by a distance of 10 meters. Blue dots represent the tag position while walking around the perimeter of the triangle and pausing at the position of the orange dots. While these tests were encouraging, we have discovered that accuracy is dependent upon: antenna orientation, and proximity to the human body and to earth. There are a number of avenues to pursue once feasibility of this technology is verified including optimization of the antenna and the radio signal characteristics. Once the anchors have detected that a pedestrian has veered out of the crosswalk, this information can be easily communicate back to the traffic controller via Ethernet over Powerline communication. However, providing navigational feedback to the pedestrian in a suitable form will require more research. Contact: James Frenzel, jfrenzel@uidaho.edu; Denise Bauer, dbauer@uidaho.edu PEDESTRIAN TRACKING USING ULTRA-WIDEBAND RADIO 9 2014-2015 Annual Report