In recent years,with the acceleration of urbanization,urban building fires are frequent,which causing serious casualties and property damage.People are often trapped in building fires,and rapid rescue of trapped people is very critical.The complex structure of large high-rise buildings and the dynamic and uncertainty of fire spread bring great difficulties to firefighters’ rescue dispatch and also pose a great threat to firefighters’ own safety.With the development of robotics,firefighting robots are increasingly used in fire rescue,which can greatly help to improve rescue capability and efficiency.Firefighters and firefighting robots have their own characteristics and are complementary in many scenarios.However,there is still limited research on the collaborative rescue of firefighters and firefighting robots.To address the above situation,this thesis investigates the collaborative rescue problem of firefighters and firefighting robots in building fire scenarios,proposes a human-robot collaborative path planning problem with the goal of minimizing rescue time,designs an ant colony optimization-based problem solving algorithm,and validates the performance on test instances.The main work of this thesis contains;1.Research on the human-robot collaborative optimal scheduling problem of single-objective rescue in building fire scenarios,taking into account the dynamic changes of the fire in building fire scenarios as an important factor affecting the rescue,and establishing a mathematical model of the problem,with the optimization objective of the model being to minimize the firefighters’ arrival at the location where the trapped people are located.In order to solve this problem more effectively,this thesis proposes a DP-ACO algorithm using a mixture of dynamic programming and ant colony optimization.The algorithm uses the ant colony optimization to solve the rescue route of the firefighters and dynamic programming to solve the rescue route of the firefighting robot so that it can reach the location of the route where it needs to advance in concert with the firefighters and wait for them as soon as possible.In the experiments conducted in 14 examples based on real buildings abstraction,the human-robot collaborative rescue is more efficient than the firefighter rescue alone,and the DP-ACO algorithm has more advantages over other mainstream algorithms in terms of solution performance.2.Research on the human-machine collaborative optimal scheduling problem of multi-objective rescue in building fire scenarios,multiple firefighters and multiple firefighting robots are needed to carry out cooperative rescue,the objective function of minimizing the average time for all trapped persons to be rescued.The DP-ACO algorithm is extended and applied to the solution of the multi-objective rescue problem by path extension.Experiments are conducted on eight different building structures with different trapped personnel locations,and the results also show that the DP-ACO algorithm outperforms other mainstream algorithms in solving this problem. |