Cooperative Control of Autonomous Unmanned Vehicles for Wildfire Management

点击翻译按钮获取中文摘要

Wildfire monitoring and fighting requires near real-time information on where and how wildfires spread in order to effectively plan and execute emergency response strategies. Existing satellite-based imaging methods of wildfire detection may not be sufficient enough to provide the required real time information on the spread of dynamic fire fronts. Also, the fire situation inspection through piloted aviation may put pilots at risk, or the operations maybe limited by the resources or weather conditions. With recent developments in unmanned vehicle systems (UVS) technology, researchers around the world have been exploring the use of unmanned vehicles, often including both unmanned aerial vehicles (UAVs) and ground vehicles (UGVs), to provide real-time wildfire monitoring. In this proposed research, we aim at developing a novel strategy for online mapping of dynamic fire fronts by flying a fleet of UAVs in formation and in coordination with UGVs. In addition, UGVs are deployed to perform protection tasks, such as search and rescue, under the guidance of UAVs. Expected research contributions include the online mapping (estimation) of fire front lines, autonomous flight and tracking, cooperative control and task allocation among UAVs/UGVs, as well as cooperative optimization and communication. "Forests and forest resources are integral to Canadian life. Canada has about 10% of the world's forests. Each year over the last decade, about 7400 forest fires have occurred. The total area burned varies widely from year to year, but averages about 1.9 million hectares annually." (source: Canadian Forest Service) In this proposed research program, we endeavor to make significant contributions to the wildfire research and management.