Abstract
Technical advancement has propelled the proliferation of unmanned vehicles. Out of the multiple paths between origin (O) and destination (D), the optimal O-D path should be selected in the light of travel distance, travel time, fuel cost and pollutant emissions. This paper proposes a dynamic path planning strategy based on fuzzy logic (FL) and improved ant colony optimization (ACO). Firstly, the classic ACO was improved into the rank-based ant system. The rank-based ant system works well in static environments, but cannot adapt well to dynamic environments. Considering the difficulty in accurate digitization of dynamic factors, the improved ACO was integrated with the FL into the fuzzy logic ant colony optimization (FLACO) to find the optimal path for unmanned vehicles. Finally, the FLACO, the classic ACO and the improved ACO were separately applied to find the optimal path in a road network, with a novel concept called virtual path length. The results show that the FLACO output the shortest virtual path among the three algorithms, i.e. identified the most cost-effective path. This mean the FLACO can find the most efficient and safe path for unmanned vehicles in a dynamic manner.
Highlights
Technical advancement has propelled the proliferation of unmanned vehicles [1]
In light of the above, this paper proposes the fuzzy logic ant colony optimization (FLACO), which can find the optimal paths for unmanned vehicles dynamically based on multiple indices
The elitist ant system might not converge to the global optimal path, because of the small difference between the candidate paths
Summary
Technical advancement has propelled the proliferation of unmanned vehicles [1]. These vehicles can travel safety from origin (O) to destination (D), with the aid of onboard high-precision sensors and intelligent control algorithms. Our route guidance system for unmanned vehicles mainly plans the optimal path based on the onboard positioning and navigation system and digital map, and guides the vehicle(s) to reach the destination rapidly and safely. 0 otherwise, where, τij is the pheromone concentration on edge ij; α = 2 controls the importance of τij [12], [23], [24]; tabu is a list of blocked edges (the visited nodes); parameters is the set of key parameters that affect the vehicle motions in large cities; l is the serial number of each parameter; αl adjusts the importance of parameter l; 1≤ ξijl ≤10 is the cost function of parameter l. The inputs of the FL control system include the travel distance, traffic flow and incident risk of the path selected by ant k. The workload of local pheromone update is minimized on the path with high travel distance, low traffic flow and low incident risk. Optimal path selection After m iterations, the FLACO system recommends the O-D path with the lowest cost
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
