The core of the Car Navigation ARM Core Board's high-precision positioning module's real-time performance lies in the coordinated collaboration of multiple technologies. Through key aspects such as hardware architecture design, sensor fusion, real-time operating system optimization, communication protocol adaptation, power management, anti-interference design, and algorithm optimization, the board ensures low latency and high reliability of positioning data in complex and dynamic environments.
The Car Navigation ARM Core Board utilizes a combination of high-performance processors and dedicated hardware acceleration units to provide the fundamental computing power for high-precision positioning. For example, the ARM Cortex-A series processors, with integrated floating-point units (FPUs) and the NEON multimedia instruction set, efficiently process raw GNSS observation data, IMU inertial measurement data, and wheel speedometer information. Some core boards also feature a hardware security module, which uses encryption algorithms to ensure the security of positioning data transmission and prevent data tampering or delays caused by malicious attacks.
Multi-sensor fusion is a key technical approach to ensuring real-time performance. The Car Navigation ARM Core Board synchronously collects data from the GNSS receiver, IMU, wheel speedometer, and steering wheel angle sensor via SPI, I2C, or CAN buses, and achieves centimeter-level positioning using Kalman filtering or factor graph optimization algorithms. For example, when a vehicle enters a tunnel or urban canyon, where GNSS signals are blocked, the IMU's angular velocity and acceleration data can maintain inertial navigation for 10-30 seconds, avoiding positioning interruptions. Wheel speedometer and steering wheel angle data are used to correct accumulated inertial navigation errors, ensuring continuous positioning output.
The real-time operating system (RTOS) scheduling mechanism is crucial to the responsiveness of the positioning module. Car navigation ARM core boards typically run hard real-time operating systems such as AUTOSAR CP or QNX, using preemptive scheduling and priority ceiling protocols to avoid task delays. For example, GNSS data parsing tasks can be assigned the highest priority to ensure immediate processing upon satellite signal reception. Non-real-time tasks such as map matching or path planning can be assigned a lower priority to prevent CPU resource consumption and positioning delays. Furthermore, the RTOS's memory protection mechanism prevents data conflicts between tasks, further improving system stability.
Optimizing communication protocols is directly related to the real-time performance of data transmission. Car navigation ARM core boards support high-speed bus protocols such as CAN FD, FlexRay, and Ethernet AVB to meet the needs of high-frequency positioning data transmission. For example, CAN FD offers a bandwidth of up to 5 Mbps, ten times that of the traditional CAN bus, enabling real-time transmission of raw GNSS observations and IMU data packets. Ethernet AVB, on the other hand, utilizes Time-Sensitive Networking (TSN) technology to achieve microsecond-level synchronization, ensuring timestamp alignment of multi-sensor data and providing an accurate time reference for subsequent fusion algorithms.
The power management design balances performance and power consumption through dynamic voltage and frequency scaling (DVFS). The car navigation arm core board adjusts the CPU frequency and core voltage in real time based on the positioning module's workload. For example, during static positioning, the frequency can be reduced to 200 MHz to conserve power, while increasing to 1 GHz to meet computing power requirements during high-speed dynamic positioning. Furthermore, the core board integrates a low-power standby mode. When the vehicle is turned off, the GNSS receiver maintains minimal power consumption while periodically uploading positioning information to the cloud. A hardware watchdog monitors system status to prevent positioning interruptions caused by program errors.
The anti-interference design optimizes signal stability in complex electromagnetic environments. The car navigation arm core board uses a shield to isolate electromagnetic interference between the GNSS antenna and onboard electronics, and integrates a SAW filter in the RF front end to suppress out-of-band noise. For example, when the LTE or 5G communication module is operating, the filter effectively blocks interference from cellular signals on the GNSS L1 band (1575.42MHz), ensuring uncompromised positioning accuracy. Furthermore, the core board supports multi-frequency GNSS reception. By simultaneously tracking signals in the L1, L2, or L5 bands, it further mitigates ionospheric delay errors and improves real-time positioning.
Algorithm optimization focuses on reducing computational latency and improving fusion accuracy. The car navigation arm core board runs a lightweight positioning engine that replaces floating-point operations with fixed-point arithmetic to reduce instruction cycles. It also employs a sliding window filter algorithm to update sensor error parameters in real time. For example, in tunnel scenarios, the algorithm dynamically adjusts the weighting between the IMU and wheel speedometer to quickly converge to the optimal positioning solution. In open road scenarios, GNSS data is prioritized to reduce the accumulated error of inertial navigation. Furthermore, the core board supports over-the-air (OTA) remote upgrades, enabling continuous optimization of the positioning algorithm to adapt to different driving scenarios.
