An Extended Bridge Weigh-in-Motion System without Vehicular Axles and Speed Detectors Using Nonnegative LASSO Regularization

The bridge weigh-in-motion (BWIM) technique uses the instrumented bridge on a large scale to identify the axle weight of a passing vehicle. Vehicle configurations, e.g., axle number and wheelbase, are crucial for the BWIM system, which require additional axle detectors. Free of axle (FAD) sensors are often used to obtain vehicle information, but they are only suitable for specific bridge types, such as slab-girder bridges. The concept of a virtual-axle-based algorithm, without requiring axle detectors, has been developed, and the validity of this algorithm has been verified numerically and experimentally. However, this algorithm assumes the vehicle speed as a known input, indicating that additional speed sensors/devices are still required in the BWIM system. Using this virtual-axle-based algorithm in a field test, it is found that the identification accuracy of the BWIM system is sensitive to the vehicle speed, and it shows poor recognition of vehicle configuration. To improve the recognition accuracy and remove vehicle speed detectors from the BWIM system, an extended BWIM system is proposed using the regularization technique and iterative approach. Both vehicular virtual axles and speeds are assumed in this approach. An error function based on the measured responses and theoretical ones is built to evaluate these assumed vehicle configurations and speeds. The effectiveness of the proposed approach is verified by the field tests. The results show that the proposed approach can obtain high recognition accuracy, which is close to Moses's algorithm using FAD sensors. Compared with the previous virtual-axle-based algorithm, the recognition accuracy and robustness of the proposed approach are greatly improved. The proposed approach is still challenged by real-world traffic because this paper only considers the case when a single vehicle passes over the bridge. Nevertheless, the proposed extended BWIM system shows potential practical applications as it can further reduce costs and be applicable to more bridge types.