Automotive Speed Estimation: Sensor Types and Error Characteristics from OBD-II to ADAS 1 Dept. of Electrical and Computer Engineering, Queen’s University at Kingston, ON, K7L3N9, Canada, and the Navigation and Instrumentation (NavINST) Lab at the Royal Military College of Canada, Kingston, ON, K7K7
Hany Ragab 1 , , Sidney Givigi 2 , , Aboelmagd Noureldin 1,2 · 2025
Read Automotive Speed Estimation: Sensor Types and Error Characteristics from OBD-II to ADAS 1 Dept. of Electrical and Computer Engineering, Queen’s University at Kingston, ON, K7L3N9, Canada, and the Navigation and Instrumentation (NavINST) Lab at the Royal Military College of Canada, Kingston, ON, K7K7 by Hany Ragab 1 , , Sidney Givigi 2 , , Aboelmagd Noureldin 1,2 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-00242.txt.
Introduction
Hany Ragab, Sidney Givigi, and Aboelmagd Noureldin present a paper on car speed sensing, with the head dated 2024. Navigation systems blend speed data with inertial and satellite tools. Most cars report speed from their diagnostic port, but cars use varied wheel sensors, and each type errs in its own way.
The authors survey speed sensors in standard cars and in advanced driver-help and self-driving systems. They propose a way to tell which sensor a car has and to model its errors. Road tests on three long drives in Toronto and Kingston, Canada, supplied the data. The results stress that mixing sensor types gives the best speed estimates. That matters most where satellite signals fail, such as city centers.
Catalog record
| CREATOR | Hany Ragab 1 , , Sidney Givigi 2 , , Aboelmagd Noureldin 1,2 |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 26 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-00242
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Automotive Speed Estimation: Sensor Types and Error Characteristics from OBD-II to ADAS 1 Dept. of Electrical and Computer Engineering, Queen’s University at Kingston, ON, K7L3N9, Canada, and the Navigation and Instrumentation (NavINST) Lab at the Royal Military College of Canada, Kingston, ON, K7K7B4, Canada. Emails: [email protected], [email protected] 2 School of Computing, Queen’s University at Kingston, ON, K7L2N8, Canada. Emails: {sidney.givigi, nourelda}@queensu.ca Hany Ragab 1 , , Sidney Givigi 2 , , Aboelmagd Noureldin 1,2 Abstract Modern on-road navigation systems heavily depend on integrating speed measurements with inertial navigation systems (INS) and global navigation satellite systems (GNSS). Telemetry-based applications typically source speed data from the On-Board Diagnostic II (OBD-II) system. However, the method of deriving speed, as well as the types of sensors used to measure wheel speed, differs across vehicles. These differences result in varying error characteristics that must be accounted for in navigation and autonomy applications. This paper addresses this gap by examining the diverse speed-sensing technologies employed in standard automotive systems and alternative techniques used in advanced systems designed for higher levels of autonomy, such as Advanced Driver Assistance Systems (ADAS), Autonomous Driving (AD), or surveying applications. We propose a method to identify the type of speed sensor in a vehicle and present strategies for accurately modeling its error characteristics. To validate our approach, we collected and analyzed data from three long real road trajectories conducted in urban environments in Toronto and Kingston, Ontario, Canada. The results underscore the critical role of integrating multiple sensor modalities to achieve more accurate speed estimation, thus improving automotive navigation state estimation, particularly in GNSS-denied environments. I Introduction Accurate speed estimation, particularly forward velocity, is crucial for automotive applications such as fleet management, navigation, Advanced Driver Assistance Systems (ADAS), and Autonomous Driving (AD) [ 1 ] . Vehicle tracking depends on real-time vehicular data for route optimization, while ADAS and AD require precision for functions like lane-keeping and collision avoidance. Mercedes-Benz’s DRIVE PILOT [ 2 ] emphasizes redundancy and safety, using moisture sensors in the wheel wells to monitor road wetness and backup systems to ensure reliable operation in challenging conditions. Many complementary systems, such as navigation and driver assistance solutions added post-manufacture, access essential vehicular data through the On-Board Diagnostic II (OBD-II) port [ 3 ] . As illustrated in Fig. 1 , the OBD-II interface connects to the vehicle’s primary network, typically the CAN bus, or other protocols depending on the make and model. It allows communication with subsystems such as the Electronic Control Units (ECUs), Anti-lock Braking System (ABS), Instrument Cluster, and Adaptive Cruise Control (ACC) modules, to name a few. This standardized interface provides critical parameters, including engine RPM and wheel speed, facilitating seamless integration with internal and external systems. However, the sensors and techniques used to derive speed measurements can differ significantly between land vehicles. These differences can introduce variations in error characteristics, which can profoundly impact the reliability and accuracy of downstream systems relying on this data. Fig. 1: Interconnections of Automotive Sensors and Systems via CAN. Wheel speed, in particular, plays a critical role in enhancing Inertial Navigation Systems (INS) and enabling the Reduced Inertial Sensor System (RISS), a simplified INS variant that predominantly depends on external speed inputs for operation [ 4 ] . The quality of wheel speed data is essential in GNSS-denied environments or, in the case of modern systems, signals of opportunity (SOPs)-denied conditions, where errors in proprioceptive measurements can propagate and degrade dead-reckoning performance. Furthermore, a limited understanding of sensor types, error characteristics, and data formats can compromise the accuracy and reliability of complementary systems. The key contributions of this paper are: Fig. 2: Methods for estimating the speed of a land vehicle without relying on GNSS or IMU integration. 1) Comprehensive overview of automotive speed sensing: We analyze methods for estimating vehicle speed, highlighting classical and modern sensor technologies. 2) Comparison of speed derivation processes: We investigate the speed derivation methods employed at the Original Equipment Manufacturer (OEM) level for different vehicle types. We also emphasize the distinctions between OBD-II and CAN bus protocols’ data format. 3) Real-road data collection and analysis: We utilize road test data to identify speed sensor types from OBD-II-reported messages, compare the accuracy of OBD-II-derived speed estimates with perception-based methods, and conduct an error variance analysis as a step toward achieving effective sensor fusion. The remainder of this paper is organized as follows: Section II reviews related work on ground vehicle speed sensing and its applications in navigation. Then, Section III explores OBD-II communication, data formatting, and speed derivation for various vehicle configurations, along with an overview of wheel speed sensors (WSS). Following this, Section IV presents experimental results, including sensor identification, error variance analysis, and the integration of stereo cameras for complementary speed estimation. Finally, Section V concludes the paper and outlines future research directions. II Related Work Several publicly available automotive datasets offer valuable insights for research and development; however, only a few include data collected directly through a vehicle’s proprietary diagnostic interface. Among datasets that provide speed data, high-resolution Distance Measuring Instruments (DMIs), such as those featured in the DARPA dataset [ 5 ] and the Ford Campus dataset [ 6 ] , are commonly used. While these systems deliver exceptional accuracy, their high cost and complexity often exceed the requirements of low-cost automotive navigation applications. In addition to built-in vehicular speed sensing and DMIs, alternative systems have been explored for specialized ground vehicle testing. Magnetic Rotary Encoders (MREs) [ 7 ] and Non-contact Optical Sensors [ 8 , 9 ] are examples used in extreme conditions like snow, ice, or wet surfaces, or in harsh environments with dust, moisture, extreme temperatures, and mechanical shocks. Despite their robustness and precision, these niche applications and high costs limit their practical use in everyday automotive systems. Datasets like nuScenes [ 10 ] , A2D2 [ 11 ] , and comma2k19 [ 3 ] aim to make built-in vehicular data more accessible. These datasets rely on the CAN bus network, which poses implementation challenges due to the lack of standardization across manufacturers. Each manufacturer defines unique CAN message structures and IDs, typically documented in DBC files. Consequently, interpreting data such as vehicle wheel speed from the CAN bus often requires extensive knowledge of the vehicle’s architecture or reverse engineering. In contrast, the OBD-II protocol provides a standardized and interoperable approach for extracting diagnostic data, including vehicle speed, across all OBD-II-compliant vehicles. Table I summarizes the key differences between the OBD-II protocol and CAN bus. Although the OBD-II protocol is straightforward, the diverse methods used for deriving vehicle speed from built-in sensors remain underexplored in the literature, creating significant challenges for navigation system integrators aiming to develop universally compatible, robust, and low-cost solutions. [b] TABLE I: Comparison between OBD-II protocol and CAN Bus Aspect OBD-II Standard CAN Bus ‡ Purpose Diagnostics and emissions monitoring Real-time communication between ECUs Standardization Standardized PIDs † (e.g., 0x0D ) Manufacturer-specific (i.e., CAN ID varies) Data Access Limited to subset of vehicle parameters Full access to all CAN messages Communication Model Request-Response Broadcast (continuous streaming) Speed of Access Slower, query-based Faster, real-time Physical Access OBD-II port OBD-II port or direct CAN wiring † PID: Parameter ID as defined in SAE