Trajectories Of Objects Research Articles

Investigating a Toolchain from Trajectory Recording to Resimulation

The growing variety of transportation options and increasing traffic congestion pose new challenges for road safety. As a result, there is an intensified focus on developing automated driving features and assistance systems aimed at minimizing accidents caused by human errors. The creation of these systems requires a substantial amount of testing kilometers, with estimates suggesting that around 2.1 billion kilometers would be necessary to ensure that each situation pertinent to the driving function is encountered at least once with a probability of 50%. This paper advances the microscopic simulation of traffic scenarios beyond linear patterns, utilizing the open-source environment openPASS. It addresses the research question of whether existing microscopic simulations are able to realistically represent non-linear traffic scenarios. A comprehensive toolchain integrates simulation with video recordings and laser scans. The study compares recorded traffic flow data with simulations at a T-junction, assessing the realism of vehicle models and trajectory representation. Three scenarios are analyzed, considering vehicles and pedestrians. The 3D geometry of the scene was captured with a laser scanner, enabling the mapping of recorded video data onto a geo-referenced environment. Object trajectories were extracted using an ’Regions with Convolutional Neural Networks features’ object detector. While openPASS simulated vehicle and pedestrian behaviors effectively, limitations in trajectory variability and reaction times were observed. These findings highlight the need for more realistic behavior models. This research emphasizes the necessity for improvements to accommodate complex driving behaviors and pedestrian dynamics.

New Astronomical Observatory Design for the Detection and Tracking of Satellite Objects: The Satellite Robotic Observatory (SRO)

Robotic Astronomical Observatories (RAOs) have provided very good results in different research projects in astrophysics/astronomy. Their applications in the detection, tracking, and identification of near-Earth objects have contributed to the identification of potentially dangerous objects for our security, such as near-Earth Objects (NEOs), near-Earth Asteroids (NEAs), meteors, and comets, whose trajectory changes can cause an impact on our planet. If advances in astrometry techniques (measuring the position and trajectory of Earth-orbiting objects) and photometry (variation in light curves) are considered together with the new sensors that work in the optical and near-infrared spectral ranges, a new observatory system that allows for the detection of nearby satellite objects in different spectral ranges and with better-defined optics can be developed. The present paper describes the design of a new observatory applied to the surveillance and tracking of satellites and other debris objects, the Satellite Robotic Observatory (SRO). Starting from general constraints from astronomy observatories, the design process has been determined, considering the main objectives, the necessary sensors, and several technical improvements that have contributed to a final configuration proposed for the SRO. The result is the design of a portable observatory model that can host at least two sensors to track and monitor satellite objects simultaneously.

AI System for Autonomous Vehicles

The rapid advancement of artificial intelligence (AI) has revolutionized the development of autonomous vehicles, offering transformative potential for the future of transportation. This research investigates the implementation of AI-driven algorithms in autonomous vehicles, focusing on their ability to enhance decision-making, navigation, and safety. By employing state-of-the-art machine learning models, including deep learning and reinforcement learning, the study explores how these technologies can optimize real-time processing of sensor data, environmental perception, and adaptive control mechanisms. The findings demonstrate that AI algorithms can significantly improve the accuracy of object detection, trajectory prediction, and path planning, thereby reducing the likelihood of collisions and enhancing overall road safety. A key contribution of this work is the integration of a multi-modal sensor fusion approach, combining data from LiDAR, cameras, radar, and GPS to create a comprehensive and reliable understanding of the vehicle’s surroundings. Additionally, the research highlights the role of AI in enabling autonomous vehicles to learn from vast amounts of driving data, facilitating continuous improvement and adaptability in diverse driving conditions. The implications of this study are profound, suggesting that AI-powered autonomous vehicles could lead to safer, more efficient, and environmentally sustainable transportation systems. However, the research also identifies challenges related to computational complexity, real-time decision-making, and ethical considerations in AI-driven autonomy. Future work will focus on addressing these challenges and exploring the broader societal impacts of widespread autonomous vehicle adoption.

Extracting terrain elevation information in front of the vehicle based on vehicle-mounted LiDAR in dynamic environments

Abstract Point cloud maps constructed using 3D LiDAR, are widely used for robot navigation and localization. Few studies have utilized point cloud maps to extract terrain elevationinformation in front of a vehicle, which can be used as active suspension inputs to reduce vehicle bumps. In addition, the trajectories of dynamic objects in point cloud maps and global navigation satellite system (GNSS) data loss can affect the extraction of elevation information. To solve these problems, this paper proposes a framework for extracting terrain elevation information in front of the vehicle based on vehicle-mounted LiDAR in dynamic environments. The framework consists of two modules: point cloud map construction and vehicle front terrain elevation information extraction. In the point cloud map construction module, a system for simultaneous localization and mapping (SLAM) is proposed, which is capable of building point cloud maps without GNSS. Furthermore, a dynamic descriptor-based dynamic object filtering algorithm is proposed which is applied to SLAM. Therefore, the SLAM system overcomes the influence of dynamic objects on vehicle position and attitude estimation, and there are no trajectories of dynamic objects in the point cloud maps built by the system. In the vehicle front terrain elevation information extraction module, the unscented Kalman filter is utilized to predict the vehicle position at the next moment. Based on the geometric features of the tire-ground contact area, the terrain elevation information of the tire contact area at the predicted position on the point cloud map is extracted. Experiments show that the algorithm in this paper overcomes the effect of dynamic objects and builds a vehicle point cloud map without dynamic objects under GNSS data loss, which improves the accuracy of the extraction of terrain elevation information in front of the vehicle.

A LiDAR-Based Digital Twinning Workflow for Traffic Monitoring and Simulation

Abstract. The process of ensuring efficient and safe urban transportation is closely linked to urban planning, particularly through the aspects of transportation planning. Transportation planning is a pivotal concern for urban regions worldwide, reflecting the growing need to increase mobility while ensuring safety and sustainability in densely populated areas. This research focuses on developing a novel digital-twin-based approach for micro-traffic simulation to support data-driven decision-making for increasing traffic safety through scenario planning. Leveraging the traffic data obtained through monitoring one of the busiest intersections in Sofia city, this research workflow shows the effective integration of LiDAR data and the urban digital twin concept in intelligent transportation systems (ITS). The research addresses problems related to moving object classification, trajectory analysis, and reclassification of unrecognised objects by processing the LiDAR data, pre-processed in a .osef format, thereby transforming it to make it suitable for simulation. The proposed solution for the monitoring of urban traffic is demonstrated by the usage of SUMO (Simulation of Urban MObility) for performing simulations and a Random Forest model for unrecognized object reclassification to pre-existing vehicles and pedestrian classes. The architecture of the proposed workflow can possibly be applied in other similar urban settings, providing a scalable solution for both traffic management and urban planning. The study’s results support the wider use of urban digital twin principles in ITS by highlighting the value of advanced modelling tools and high-quality data in addressing today's urban transportation challenges.

Enhancing Availability of Autonomous Shuttle Services: A Conceptual Approach toward Challenges and Opportunities

<div>Autonomous vehicle technologies have become increasingly popular over the last few years. One of their most important application is autonomous shuttle buses that could radically change public transport systems. In order to enhance the availability of shuttle service, this article outlines a series of interconnected challenges and innovative solutions to optimize the operation of autonomous shuttles based on the experience within the Shuttle Modellregion Oberfranken (SMO) project. The shuttle shall be able to work in every weather condition, including the robustness of the perception algorithm. Besides, the shuttle shall react to environmental changes, interact with other traffic participants, and ensure comfortable travel for passengers and awareness of VRUs. These challenging situations shall be solved alone or with a teleoperator’s help. Our analysis considers the basic sense–plan–act architecture for autonomous driving. Critical components like object detection, pedestrian tracking, trajectory planning, and teleoperation are analyzed, and concepts for overcoming the challenges mainly in these components are presented.</div>

The "effects of wind turbine sound on working memory" (a pre-registration plan)

Although research on the effects of wind turbine sound (WTS) over mental health factors is increasing, very little attention has been given to the potential effects of WTS over subtle cognitive processes, such as concentration or memory. The "brainwave entrainment hypothesis" is rooted in neuropsychological research demonstrating that brain oscillations are modulated by external rhythms. In addition, perceived annoyance to wind turbines seems to be a phenomenon in which both visual and psychological variables interact with acoustic stimulation. However, annoyance is a concept rarely operationalised in the literature. Our study aims to experimentally assess the effects of WTS on working memory. Working memory is recognised as a basic process, playing a crucial role in unconscious everyday processes (e.g., perceiving an object's trajectory despite input interruptions when blinking), as well as deliberate processes (e.g., carrying out arithmetic operations). By exploring variations in performance during a memory task, while being exposed to WTS samples, we could highlight acoustic characteristics of the WTS stimuli that interfere-or not-with cognitive processes. These data, combined with within-block ratings of annoyance, can provide us with index of annoyance directly related to the acoustic properties present in WTS recordings.

SketchQL: Video Moment Querying with a Visual Query Interface

Localizing video moments based on the movement patterns of objects is an important task in video analytics. Existing video analytics systems offer two types of querying interfaces based on natural language and SQL, respectively. However, both types of interfaces have major limitations. SQL-based systems require high query specification time, whereas natural language-based systems require large training datasets to achieve satisfactory retrieval accuracy. To address these limitations, we present SketchQL, a video database management system (VDBMS) for offline, exploratory video moment retrieval that is both easy to use and generalizes well across multiple video moment datasets. To improve ease-of-use, SketchQL features a visual query interface that enables users to sketch complex visual queries through intuitive drag-and-drop actions. To improve generalizability, SketchQL operates on object-tracking primitives that are reliably extracted across various datasets using pre-trained models. We present a learned similarity search algorithm for retrieving video moments closely matching the user's visual query based on object trajectories. SketchQL trains the model on a diverse dataset generated with a novel simulator, that enhances its accuracy across a wide array of datasets and queries. We evaluate SketchQL on four real-world datasets with nine queries, demonstrating its superior usability and retrieval accuracy over state-of-the-art VDBMSs.

Using posture recognition algorithms based on machine learning to identify senior health

Faced with the situation that the elderly people at home have dangerous behaviors, the study explores various aspects of motion target detection, real-time target tracking and behavioral pose recognition and classification, using behavioral poses in videos as samples. To tackle the challenges in detecting motion targets, a target detection method based on Gaussian mixture model (GMM) and four frame difference method is proposed; A tracking technique incorporating Kalman filter (KF) is investigated to trail the behavioral changes of the elderly in actual time. A seven-layer convolutional neural network (CNN) is constructed to face the problem of inaccurate behavioral pose recognition. Through relevant experimental analyses, the outcomes show that the increased GMM detection way has a complete profile and the accuracy is significantly improved. The KF target tracking technique can trail the object trajectory in actual time and steadily, with the smallest trailing error value of 0.19. The classification accuracy of the CNN pose recognition model is 95.87%, and the pose classification time is 27 seconds. Its performance is superior to the mean shift algorithm, particle filter algorithm, and Cam Shift algorithm in all aspects. When applied in practice, it can accurately identify whether the elderly’s behavior is abnormal and ensure their daily health.

A three-dimensional trajectory measurement system for quasi-one-dimensional high-speed moving objects

Abstract The three-dimensional (3D) trajectory measurement of a quasi–one-dimensional (quasi-1D) high-speed moving object often cannot achieve the optimizations of both the measurement range and the object’s resolution at the same time. In this paper, a 3D trajectory measurement system for quasi-1D high-speed moving objects is proposed. The system was composed of a single camera and a galvo mirror. The posture of the virtual camera is changed by adjusting the angle of the galvo mirror, thereby achieving the tracking and imaging of a quasi-1D high-speed moving object. For a background with complex textures, a measurement method based on the reconstructed homography matrix is proposed. The extrinsic matrix of the virtual camera is estimated based on the feature points on the background, and the equation of the object trajectory’s plane is estimated based on the feature points on the object. Then, by combining the extrinsic matrix and the equation of the plane, a new homography matrix is reconstructed, realizing the measurement of the 3D coordinates of the object. For a background with weak textures, a method based on the deflection angle of the galvo mirror is proposed for the calibration of the extrinsic matrix of the virtual camera. This method can calculate the extrinsic matrix of the virtual camera based on the distance from the camera to the center point of the galvo mirror and the deflection angle of the galvo mirror. The experimental results show that the proposed system is able to increase the field of view by 430%. For complex-texture backgrounds, the relative error of the system is 0.79%, and for weak-texture backgrounds, the relative error is 4.65%.

TrajectoryNAS: A Neural Architecture Search for Trajectory Prediction.

Autonomous driving systems are a rapidly evolving technology. Trajectory prediction is a critical component of autonomous driving systems that enables safe navigation by anticipating the movement of surrounding objects. Lidar point-cloud data provide a 3D view of solid objects surrounding the ego-vehicle. Hence, trajectory prediction using Lidar point-cloud data performs better than 2D RGB cameras due to providing the distance between the target object and the ego-vehicle. However, processing point-cloud data is a costly and complicated process, and state-of-the-art 3D trajectory predictions using point-cloud data suffer from slow and erroneous predictions. State-of-the-art trajectory prediction approaches suffer from handcrafted and inefficient architectures, which can lead to low accuracy and suboptimal inference times. Neural architecture search (NAS) is a method proposed to optimize neural network models by using search algorithms to redesign architectures based on their performance and runtime. This paper introduces TrajectoryNAS, a novel neural architecture search (NAS) method designed to develop an efficient and more accurate LiDAR-based trajectory prediction model for predicting the trajectories of objects surrounding the ego vehicle. TrajectoryNAS systematically optimizes the architecture of an end-to-end trajectory prediction algorithm, incorporating all stacked components that are prerequisites for trajectory prediction, including object detection and object tracking, using metaheuristic algorithms. This approach addresses the neural architecture designs in each component of trajectory prediction, considering accuracy loss and the associated overhead latency. Our method introduces a novel multi-objective energy function that integrates accuracy and efficiency metrics, enabling the creation of a model that significantly outperforms existing approaches. Through empirical studies, TrajectoryNAS demonstrates its effectiveness in enhancing the performance of autonomous driving systems, marking a significant advancement in the field. Experimental results reveal that TrajcetoryNAS yields a minimum of 4.8 higger accuracy and 1.1* lower latency over competing methods on the NuScenes dataset.

Establishing requirements for lunar and cislunar orbital debris tracking

There is a need to formally coordinate lunar and cislunar space traffic management to mitigate the risk of collisions with micrometeoroids and orbital debris in this space domain. To control this hazard, it will be critical to develop a high-fidelity orbital debris catalog. This catalog will be maintained by monitoring and propagating the trajectory of objects. The need to perform a debris avoidance maneuver for collision avoidance will depend on the fidelity of the propagated debris trajectory. A larger uncertainty (magnitude as a function of time) will require a larger maneuver and a higher likelihood of the need to perform this maneuver. This study assesses debris avoidance maneuvers and corresponding corrective actions to recover the desired mission trajectory as a way to evaluate the desired capability of tracking of objects in lunar and cislunar orbit. The baseline Earth-to-Moon trajectory was that of the Artemis I mission. Typical conjunction assessments were postulated at 100,000 km from Earth's center (approximately a quarter of the way to the moon) and at lunar orbit insertion (LOI). For these two cases, the required debris avoidance maneuver (and hence impact to the mission) is tied to the uncertainty in orbital debris tracking. The study provides a methodology and baseline inputs to establish future requirements of debris tracking in lunar and cislunar orbits. It also reinforces the importance of long-term sustainability for lunar missions, specifically preventing the generation of orbital debris.

Near-Infrared Response Organic Synaptic Transistor for Dynamic Trace Extraction.

The development of neuromorphic hardware capable of detecting and recognizing moving targets through an in-sensor computing strategy is considered to be an important component of the construction of edge computing systems with distributed computation. In addition to responsiveness to visible light, the implementation of neuromorphic hardware should also demonstrate the ability to sense and process nonvisible light, which is essential for tracking target object trajectories in specialized environments. In this work, we fabricated an organic synaptic transistor with a near-infrared (NIR) response by incorporating doped LaF3: Yb/Ho upconversion quantum dots (UCQDs) into the channel of a Poly3-hexylthiophene (P3HT)-based organic field effect transistor (FET), serving as charge trapping and infrared sensing sites. The obtained synaptic transistor not only replicates common synaptic behaviors when exposed to NIR illumination but also demonstrates potential applications for the dynamic trajectory recognition of animals in the dark. Compared to other monitoring technologies, P3HT transistors doped with LaF3: Yb/Ho UCQDs exhibit distinct advantages, including a NIR response, high-efficiency computing, and sensitivity, which provide an experimental foundation and a design reference for the development of next-generation intelligent dynamic image recognition systems.

Smart3DMOT: Smart cascade 3D MOT tracking strategy with motion and appearance association

3D multi-object tracking (MOT) is pivotal for associating the trajectories of objects in autonomous driving. The integration of image and point cloud data in 3D MOT algorithms has emerged as a research hotspot, aiming to achieve both high precision and efficiency tracking. In this paper, we present Smart3DMOT, an novel method that leverages multi-modal detection data and incorporates a cascade matching strategy, intelligently utilizing motion and multi-modal appearance affinity for data association across frames. Our algorithm introduces two association models: the strong motion association model (SM-AM) and the multi-modal fusion feature association model (MMFF-AM), tailored to different tracking scenarios. The SM-AM robustly associates objects by enhancing motion similarity, while the MMFF-AM integrates multi-modal fusion features from images and point clouds to generate appearance similarity between objects and combine motion cues for joint reasoning. Finally, we propose a cascade tracking strategy with a “spatial tracking state confirmer” (STSC). The SM-AM and MMFF-AM models are performed staged matching facilitated by STSC. This intelligent implementation allows SM-AM or MMFF-AM to tackle instances with different tracking difficulties, maintaining the operational efficiency and tracking accuracy of trade-offs. Our approach demonstrates the effectiveness of the proposed modules through comprehensive ablation experiments and the results obtained in the KITTI and nuScenes benchmark show that our method attains state-of-the-art performance in 3D MOT.

Effects of multisensory stimulation on infants' learning of object pattern and trajectory.

This study investigated whether infants encode better the features of a briefly occluded object if its movements are specified simultaneously by vision and audition than if they are not (data collected: 2017-2019). Experiment 1 showed that 10-month-old infants (N = 39, 22 females, White-English) notice changes in the visual pattern on the object irrespective of the stimulation received (spatiotemporally congruent audio-visual stimulation, incongruent stimulation, or visual-only; = .53). Experiment 2 (N = 72, 36 female) found similar results in 6-month-olds (Test Block 1, = .13), but not 4-month-olds. Experiment 3 replicated this finding with another group of 6-month-olds (N = 42, 21 females) and showed that congruent stimulation enables infants to detect changes in object trajectory (d = 0.56) in addition to object pattern (d = 1.15), whereas incongruent stimulation hinders performance.

The cat is out of the bag: the effect of virtual characters and scale cues on physics coherence

According to previous research, humans are exceptionally skilled at distinguishing earth-congruent object trajectories from earth-discrepant ones. Inconsistent scale cues in Virtual Reality (VR), however, are shown to easily confuse VR users’ scale, size, and distance estimations. This paper describes the results of two studies, with 40 participants each, which tested the effects of self-scaling, and size cues from active virtual characters, on perceived physics coherence (i.e., subjective realism). In the first study (Study A), a humanoid robot the size of a doll manipulated household objects to demonstrate their rigid body dynamics. The demonstration sequence was performed twice, once using a correct approximation of physics, and once using an incorrect one in which gravity was simulated similarly to as if the doll-sized robot was human-sized and the surrounding room was enlarged. The participants observed both demonstrations, once while standing at the normal scale and once at a reduced scale, similar to that of the humanoid robot. The second study (Study B) was similar, except that the virtual character demonstrating the physics was a regular-sized cat. Our preregistered hypotheses predicted that participants would consider the correct approximation of physics as better matching their expectations at normal scale, and the incorrect one as better matching their expectations at the reduced scale. However, only the second of these hypotheses was supported. According to our exploratory analyses, the participant’s own scale was a poor predictor of physics preference, and instead there was a significant effect regarding the virtual character’s identity. Participants observing the virtual cat were about eight times more likely to select the realistic physics model compared to those observing the humanoid robot. The results indicate that familiar cues tied to virtual character identity overrode any potential effects related to changes in the participants’ own scale.

Estimating Wind Speeds in Tornadoes Using Debris Trajectories of Large Compact Objects

Abstract Currently, the enhanced Fujita scale does not consider the wind-induced movement of various large compact objects such as vehicles, construction equipment, and farming equipment/haybales that are often found in postevent damage surveys. One reason for this is that modeling debris in tornadoes comes with considerable uncertainties since there are many parameters to determine, leading to difficulties in using trajectories to analyze wind speeds of tornadoes. This paper aims to develop a forensic tool using analytical tornado models to estimate lofting wind speeds based on trajectories of large compact objects. This is accomplished by implementing a Monte Carlo simulation to randomly select the parameters and plotting cumulative distribution functions showing the likelihood of lofting at each wind speed. After analyzing the debris lofting from several documented tornadoes in Canada, the results indicate that the method provides threshold lofting wind speeds that are similar to the estimated speeds given by other methods. However, the introduction of trajectories produces estimated lofting wind speeds that are higher than the EF-scale rating given from the ground survey assessment based on structural damage. Further studies will be required to better understand these differences.

SketchQL Demonstration: Zero-Shot Video Moment Querying with Sketches

In this paper, we will present SketchQL, a video database management system (VDBMS) for retrieving video moments with a sketch-based query interface. This novel interface allows users to specify object trajectory events with simple mouse drag-and-drop operations. Users can use trajectories of single objects as building blocks to compose complex events. Using a pre-trained model that encodes trajectory similarity, SketchQL achieves zero-shot video moments retrieval by performing similarity searches over the video to identify clips that are the most similar to the visual query. In this demonstration, we introduce the graphic user interface of SketchQL and detail its functionalities and interaction mechanisms. We also demonstrate the end-to-end usage of SketchQL from query composition to video moments retrieval using real-world scenarios.

MotionTrack: Learning motion predictor for multiple object tracking

Significant progress has been achieved in multi-object tracking (MOT) through the evolution of detection and re-identification (ReID) techniques. Despite these advancements, accurately tracking objects in scenarios with homogeneous appearance and heterogeneous motion remains a challenge. This challenge arises from two main factors: the insufficient discriminability of ReID features and the predominant utilization of linear motion models in MOT. In this context, we introduce a novel motion-based tracker, MotionTrack, centered around a learnable motion predictor that relies solely on object trajectory information. This predictor comprehensively integrates two levels of granularity in motion features to enhance the modeling of temporal dynamics and facilitate precise future motion prediction for individual objects. Specifically, the proposed approach adopts a self-attention mechanism to capture token-level information and a Dynamic MLP layer to model channel-level features. MotionTrack is a simple, online tracking approach. Our experimental results demonstrate that MotionTrack yields state-of-the-art performance on datasets such as Dancetrack and SportsMOT, characterized by highly complex object motion.

LaMMOn: language model combined graph neural network for multi-target multi-camera tracking in online scenarios

Multi-target multi-camera tracking is crucial to intelligent transportation systems. Numerous recent studies have been undertaken to address this issue. Nevertheless, using the approaches in real-world situations is challenging due to the scarcity of publicly available data and the laborious process of manually annotating the new dataset and creating a tailored rule-based matching system for each camera scenario. To address this issue, we present a novel solution termed LaMMOn, an end-to-end transformer and graph neural network-based multi-camera tracking model. LaMMOn consists of three main modules: (1) Language Model Detection (LMD) for object detection; (2) Language and Graph Model Association module (LGMA) for object tracking and trajectory clustering; (3) Text-to-embedding module (T2E) that overcome the problem of data limitation by synthesizing the object embedding from defined texts. LaMMOn can be run online in real-time scenarios and achieve a competitive result on many datasets, e.g., CityFlow (HOTA 76.46%), I24 (HOTA 25.7%), and TrackCUIP (HOTA 80.94%) with an acceptable FPS (from 12.20 to 13.37) for an online application.