Brain Storm Optimized Swarm Collaboration for Bus Scheduling

Smart traffic control is a promising technology to exploit the available transportation resource and relieve traffic congestion in smart cities. This paper considers the bus scheduling problem, where a swarm of unmanned buses is supposed to intelligently accomplish tasks such as on-demand dynamic passenger pick-up or drop-off. Inspired by the dynamic window approach (DWA), To analyze this problem, the dynamic window approach (DWA) is adapted for the bus swarm scenario, and then an artificial sparrow model of neighborhood collaboration strategy is employed. Particularly, swarm coordination is realized by using a normalized Finsler geometric measured area instead of traditional Euclidean norm to coordinate the service area coverage among buses, which can help improve the local path planning of the bus swarm. The brain storm optimization (BSO) is further applied to enable the dynamic optimization and adjustment of bus swarm through analyzing the swarm motion, distribution, and the speed at which the individual reaches the target. Numerical results corroborate the effectiveness of the proposed detection and decision mechanism.