Search Results (7)

To avoid collisions and ensure driving safety, comfort, and efficiency, in this study, we propose a trajectory planning strategy for intelligent vehicles navigating curvy road scenarios. This strategy is based on a dynamic spatio-temporal corridor search. First, an obstacle space expansion module is constructed using a critical safety distance model to generate a searchable spatio-temporal corridor. Next, a dynamic step expansion is performed to improve the traditional hybrid A* search algorithm by the discretization of front-wheel steering angles and acceleration. The bisection method is applied to iteratively optimize the child nodes at each step, and the child node with the lowest cost is selected as the rough search node. Subsequently, a locally weighted dual-regression fitting algorithm is employed for segment trajectory fitting, and the optimal trajectory is generated. Finally, the performance of the proposed trajectory planning strategy is validated on the Carla simulation platform. The results show the effectiveness and efficiency of our strategy in three typical scenarios. Full article

Autonomous driving is a complex problem that requires intelligent decision making, and it has recently garnered significant interest due to its potential advantages in convenience and safety. In autonomous driving, conventional path planning to reach a destination is a time-consuming challenge. Therefore, learning-based approaches have been successfully applied to the controller or decision-making aspects of autonomous driving. However, these methods often lack explainability, as passengers cannot discern where the vehicle is headed. Additionally, most experiments primarily focus on highway scenarios, which do not effectively represent road curvature. To address these issues, we propose a reinforcement-learning-based trajectory learning in the Frenet frame (RLTF), which involves learning trajectories in the Frenet frame. Learning trajectories enable the consideration of future states and enhance explainability. We demonstrate that RLTF achieves a 100% success rate in the simulation environment, considering future states on curvy roads with continuous obstacles while overcoming issues associated with the Frenet frame. Full article

In this paper, a novel method is demonstrated to sustain vehicle-to-vehicle (V2V) communication on curvy roads via the arrangement of the lateral position of a self-angle-adjustable mirror–reflective road sign (SAAMRS) and light-direction-sensing wide-angle complementary photodiodes (CPDs). Visible light communication (VLC) between vehicles attracts attention as a complementary technology to radio-frequency-based (RF-based) communication technologies due to its wide, license-free spectrum and immunity to interferences. However, V2V VLC may be interrupted on curvy roads due to the limited field of view (FOV) of the receiver or the line of sight (LOS) being interrupted. To solve this problem, an experiment was developed using an SAAMRS along with wide-angle light-direction-sensing CPDs that used a precise peak detection (PPD) method to sustain communication between vehicles in dynamic environments by rerouting the incident light with the highest signal intensity level to the receiver vehicle on curvy roads. We also used real images of curvy roads simulated as polynomials to calculate the necessary rotation angles for the SAAMRS and regions where communication exist. Our experimental results overlapped almost completely with our simulations, with small errors of approximately 4.8% and 4.4% for the SAAMRS angle and communication region, respectively. Full article

Unlike car navigation, where almost all vehicles can traverse every route, one route might not be optimal or even suitable for all pedestrians. Route geometry information, including tortuosity, twists and turns along roads, junctions, and road slopes, among others, matters a great deal for specific types of pedestrians, particularly those with limited mobility, such as wheelchair users and older adults. Offering practical routing services to these users requires that pedestrian navigation systems provide further information on route geometry. Therefore, this article proposes a novel method for extracting and analyzing the geometry properties of the shortest pedestrian paths, with a focus on open geospatial data across four aspects: (a) similarity, (b) route curviness, (c) road turns and intersections, and (d) road gradients. Deriving from the Hausdorff distance, a metric called the “dissimilarity ratio” was developed, allowing us to determine whether pairs of routes show any tendencies to be similar to each other. Using the “sinuosity index”, a segment-based technique quantified the route curviness based on the number and degree of the road turns along the route. Moreover, relying upon open elevation data, the road gradients were extracted to identify routes offering smoother motion and better accessibility. Lastly, the road turns and intersections were investigated as pedestrian convenience and safety indicators. A local government area of Greater Sydney in Australia was chosen as the case study. The analysis was implemented on OpenStreetMap (OSM) shortest pedestrian paths against Google Maps as a benchmark for real-world commercial applications. The similarity analysis indicated that over 90% of OSM routes were identical or roughly similar to Google Maps. In addition, while Spearman’s rank correlation showed a direct relationship between route curviness and route length, $r_S\left(758\right)$ = 0.92, p < 0.001, OSM, on average, witnessed more tortuous routes and, consequently, shorter straight roads between turns. However, OSM routes could be more suitable for pedestrians when the frequency of intersections and road slopes are at the center of attention. Finally, the devised metrics in this study, including the dissimilarity ratio and sinuosity index, showed their practicability in translating raw values into meaningful indicators. Full article

Unmanned aerial vehicles (UAVs) have significant abilities for automatic detection and mapping of urban surface materials due to their high resolution. It requires a massive quantity of data to understand the ground material properties. In recent days, computer vision based approaches for intelligent transportation systems (ITS) have gained considerable interest among research communities and business people. Road classification using remote sensing images plays a vital role in urban planning. It remains challenging because of scene complexity, fluctuating road structures, and inappropriate illumination circumstances. The design of intelligent models and other machine learning (ML) approaches for road classification has yet to be further explored. In this aspect, this paper presents a metaheuristics optimization with deep autoencoder enabled road classification model (MODAE-RCM). The presented MODAE-RCM technique mainly focuses on the classification of roads into five types, namely wet, ice, rough, dry, and curvy roads. In order to accomplish this, the presented MODAE-RCM technique exploits modified fruit fly optimization (MFFO) with neural architectural search network (NASNet) for feature extraction. In order to classify roads, an interactive search algorithm (ISA) with a DAE model is used. The exploitation of metaheuristic hyperparameter optimizers helps to improve the classification results. The experimental validation of the MODAE-RCM technique was tested by employing a dataset comprising five road types. The simulation analysis highlighted the superior outcomes of the MODAE-RCM approach to other existing techniques. Full article

Autonomous vehicles are expected to take complete control of the driving process, enabling the former drivers to act as passengers only. This could lead to increased sickness as they can be engaged in tasks other than driving. Adopting different sickness mitigation techniques gives us unique types of motion sickness in autonomous vehicles to be studied. In this paper, we report on a study where we explored the possibilities of assessing motion sickness with electrogastrography (EGG), a non-invasive method used to measure the myoelectric activity of the stomach, and its potential usage in autonomous vehicles (AVs). The study was conducted in a high-fidelity driving simulator with a virtual reality (VR) headset. There separate EGG measurements were performed: before, during and after the driving AV simulation video in VR. During the driving, the participants encountered two driving environments: a straight and less dynamic highway road and a highly dynamic and curvy countryside road. The EGG signal was recorded with a proprietary 3-channel recording device and Ag/AgCl cutaneous electrodes. In addition, participants were asked to signalize whenever they felt uncomfortable and nauseated by pressing a special button. After the drive they completed also the Simulator Sickness Questionnaire (SSQ) and reported on their overall subjective perception of sickness symptoms. The EGG results showed a significant increase of the dominant frequency (DF) and the percentage of the high power spectrum density (FSD) as well as a significant decrease of the power spectrum density Crest factor (CF) during the AV simulation. The vast majority of participants reported nausea during more dynamic conditions, accompanied by an increase in the amplitude and the RMS value of EGG. Reported nausea occurred simultaneously with the increase in EGG amplitude. Based on the results, we conclude that EGG could be used for assessment of motion sickness in autonomous vehicles. DF, CF and FSD can be used as overall sickness indicators, while the relative increase in amplitude of EGG signal and duration of that increase can be used as short-term sickness indicators where the driving environment may affect the driver. Full article

Extracting road segments by averaging GPS trajectories is very challenging. Most existing averaging strategies suffer from high complexity, poor accuracy, or both. For example, finding the optimal mean for a set of sequences is known to be NP-hard, whereas using Medoid compromises the quality. In this paper, we introduce three extremely fast and practical methods to extract the road segment by averaging GPS trajectories. The methods first analyze three descriptors and then use either a simple linear model or a more complex curvy model depending on an angle criterion. The results provide equal or better accuracy than the best existing methods while being very fast, and are therefore suitable for real-time processing. The proposed method takes only 0.7% of the computing time of the best-tested baseline method, and the accuracy is also slightly better (62.2% vs. 61.7%). Full article