Potholes on road surfaces pose a serious hazard to vehicles and passengers due to the difficulty detecting them and the short response time. Therefore, many government agencies are applying various pothole-detection algorithms for road maintenance. However, current methods based on object detection are unclear in terms of real-time detection when using low-spec hardware systems. In this study, the SPFPN-YOLOv4 tiny was developed by combining spatial pyramid pooling and feature pyramid network with CSPDarknet53-tiny. A total of 2665 datasets were obtained via data augmentation, such as gamma regulation, horizontal flip, and scaling to compensate for the lack of data, and were divided into training, validation, and test of 70%, 20%, and 10% ratios, respectively. As a result of the comparison of YOLOv2, YOLOv3, YOLOv4 tiny, and SPFPN-YOLOv4 tiny, the SPFPN-YOLOv4 tiny showed approximately 2–5% performance improvement in the mean average precision (intersection over union = 0.5). In addition, the risk assessment based on the proposed SPFPN-YOLOv4 tiny was calculated by comparing the tire contact patch size with pothole size by applying the pinhole camera and distance estimation equation. In conclusion, we developed an end-to-end algorithm that can detect potholes and classify the risks in real-time using 2D pothole images.
In the present study, a novel approach to estimating the efficiency of roller chain power transmission systems is proposed based on sliding friction losses and damping force. The dynamics model is taken into account between chain links with lateral offset owing to the derailleur system. Frictional losses were calculated according to Coulomb’s law of friction, and the damping force was dependent on the damping coefficient. The effects of rotational speed, load, derailleur system, and damping coefficient on transmission efficiency were analyzed. The test stand of the roller chain power transmission system was set up to verify the estimated efficiency, and the results showed a good correlation, demonstrating the validity of the chain power transmission efficiency estimation.
In this study, a method for estimating the efficiency of electric bicycle power train systems consisting of typical components, such as an electric motor, gears, sprockets, and chains is presented. In order to calculate the efficiency of a power train system, the relationship between the drive motor torque and the road-load that is exerted on the rear wheel was derived, considering kinematic inertia effects and friction losses between power transmission elements. Among the factors that influence efficiency, it was found that friction losses play a dominant role, while the effects of inertia are insignificant. The factors that influence the efficiency of electric bicycles due to friction losses, such as the transmission efficiency of the chain system and the bearing in the sprocket and wheel, were quantified. To validate the proposed efficiency calculation procedure, an experimental electric bicycle was used, in which the driving torque and road-load could be quantitatively assessed, and the actual efficiency was measured on a chassis dynamometer. It is shown that for a given motor torque, a measured and estimated dynamometer torque obtained by the proposed method exhibits a good correlation, and the transmission efficiency of each component was quantified. This method provides a practical and accurate means to calculate the drive train efficiency of electric bicycles at the design stage to improve the efficiency of electric bicycles.
5 December 2023 in IET Intelligent Transport Systems
Scenario‐based test methods are employed to assess the safety and performance of autonomous vehicles. The analytic hierarchy process (AHP) method is a common assessment method for determining the criticality of test scenarios. However, the AHP method is subjective and less reproducible when performed by different persons, as the elements of pairwise comparison values that are directly linked to the outcome must be assigned by the person involved. This paper proposes a novel AHP method that automatically generates pairwise comparison values by optimizing the correlation between performance metrics and risk of test scenarios by simulation. Performance metrics are defined as the minimum relative distances and corresponding relative velocities between vehicles, and the risk of the test scenario is determined by the pairwise comparison values of AHP. The novel AHP method was evaluated using a cut‐in scenario. The results showed that the minimum relative distance and the risk determined by the novel AHP method achieved a better correlation coefficient of −0.96, which is better than the conventional AHP of −0.828 and Fuzzy AHP of −0.824. These results suggest that the criticality of the test scenarios determined by the novel AHP method can more accurately reflect real‐world driving environments.
The selection of an appropriate method for modeling vehicle dynamics heavily depends on the application. Due to the absence of human intervention, the demand for an accurate and real-time model of vehicle dynamics for intelligent control increases for autonomous vehicles. This paper develops a multibody vehicle model for longitudinal-vertical dynamics applicable to advanced driver assistance (ADAS) applications. The dynamic properties of the chassis, suspension, and tires are considered and modeled, which results in accurate vehicle dynamics and states. Unlike the vehicle dynamics models built into commercial software packages, such as ADAMS and CarSim, the proposed nonlinear dynamics model poses the equations of motion using a subset of relative coordinates. Therefore, the real-time simulation is conducted to improve riding performance and transportation safety. First, a vehicle system is modeled using a semi-recursive multibody dynamics formulation, and the vehicle kinematics and dynamics are accurately calculated using the system tree-topology. Second, a fork-arm removal technique based on the rod-removal technique is proposed to reduce the number of bodies, relative coordinates, and equations constrained by loop-closure. This increase the computational efficiency even further. Third, the dynamic simulations of the vehicle are performed on bumpy and sloping roads. The accuracy and efficiency of the numerical results are compared to the reference data. The comparative results demonstrate that the proposed vehicle model is effective. This efficient model can be utilized for the intelligent control of vehicle ADAS applications, such as forward collision avoidance, adaptive cruise control, and platooning.
1 April 2021 in Transaction of the Korean Society of Automotive Engineers