Small underwater unmanned platforms such as underwater gliders and profile buoys are characterized by small size, low speed and weak load capacity. According to these characteristics, an underwater acoustic target detection system with small size and low power consumption was developed, which uses a single large depth combined co-vibrating vector hydrophone as acoustic sensor, uses DSP+FPGA+EMMC architecture to build the hardware platform of the signal processing system. Then the estimated DOA relative to the true north is calculated by windowed histogram statistics for single vector hydrophone on moving platform using the real-time attitude Euler angles of the system which is obtained by the attitude measurement sensor. The system was carried out on an underwater glider for sea tests, and test results show that the system can work stably at the depth of 1000 m and the detection distance to the surface ship with a speed of 14 kn can reach 14.4 km, which proves that the system is suitable for underwater acoustic target detection on small underwater unmanned platforms.