Target maneuver recognition is crucial for air situational awareness and airspace management. However, existing algorithms require high feature dimensions and cannot accurately identify target states in real-time noisy environments. In response to these challenges, we propose an online intelligent maneuver recognition algorithm based on 1-Dimensional Convolutional Neural Network-Gated Recurrent Unit (1DCNN-GRU). Firstly, establish a trajectory database using the three-dimensional coordinates of the target for network training; Secondly, design the 1DCNN-GRU algorithm, extract high-dimensional feature using 1DCNN, and capture temporal dependencies using GRU to jointly achieve maneuver pattern recognition; Then, compared with conventional algorithms, the results show that the proposed algorithm can achieve a recognition accuracy of 91.1%, with an average running time of only 2.65 ms for a single set of data, demonstrating high accuracy and real-time performance; Finally, simulation verification shows that the proposed algorithm has good tolerance to noise and can effectively achieve online maneuvering recognition for complex moving targets.