The development of airborne embedded computer systems is characterized by rapid speed, high integration, and complex technology. Generally, they adopt a hardware architecture consisting of a processor, a central control unit, memory, and other hardware interfaces. Embedded computers usually have multiple types of reset designs to ensure the initialization of hardware resources when powered on or in specific operating states. The reset types typically include power-on reset, software reset generated by applications, external discrete quantity reset, watchdog reset, and other types of reset. Abnormal reset in airborne embedded computer systems, especially in the air, has the characteristics of low recurrence probability, poor traceability, and high difficulty in troubleshooting. Therefore, a technology that can quickly identify the type of abnormal reset and support the investigation of the cause of the abnormal reset is needed. This paper provides a method for locating abnormal reset in embedded computer systems. This method can accurately capture the type of reset when an abnormal reset occurs and record it, providing support for further locating the cause of the abnormal reset.