Aiming at the dynamic network topology environment of small unmanned aerial vehicle (UAV) ad hoc networks, a low-overhead routing algorithm based on OLSR (LS-OLSR) for small UAV ad hoc networks was proposed to solve the problem of high routing overhead of traditional optimized link state routing (OLSR) protocol. The algorithm introduced the adaptive adjustment mechanism of HELLO message and topology control (TC) message sending interval and the active update strategy of MPR willingness value, and used the real-time network status to adaptively adjust the HELLO message sending interval and TC message sending interval to reduce unnecessary control message forwarding. At the same time, the node MPR willingness value was adaptively adjusted according to the node receiving power and data transmission delay to improve the stability of the data forwarding node. The simulation results show that in the small UAV ad hoc network model, the improved routing algorithm does not increase the network end-to-end delay, and the routing overhead of the ad hoc network has decreased by about 7%, and the package loss rate is reduced by about 17%. The overall reduction of the delay of network data transmission has improved the efficiency and performance of the ad hoc network system. Therefore, LS-OLSR, as an OLSR protocol for the optimization of the ad hoc network environment of small drones, makes up for the deficiencies of traditional algorithms, and provides new solutions for high-efficiency communication of drones to network.