Trajectory Prediction Research Articles

A Trajectory Prediction Method for Reentry Glide Vehicles via Adaptive Cost Function

This paper proposes a trajectory prediction method via the adaptive cost function to address the difficulties in inferring the attack intention and maneuver mode, as well as the accumulation of prediction error during the trajectory prediction of reentry glide vehicles. Firstly, the vehicle guidance task is divided into two distinct categories: conventional guidance and no-fly zone avoidance guidance. A task-matched time-varying parameter prediction model set is then constructed. Secondly, taking into account the maneuverability, guidance intent, and battlefield situation of the vehicle, an adaptive intent cost function adapted to the guidance task is proposed, which avoids the estimation failure problem caused by manually setting cost coefficients in traditional methods. Finally, long-term trajectory prediction of vehicles is achieved using Bayesian theory to infer the attack intent and parametric model with the maximum a posteriori probability. The results of the simulations demonstrate that the proposed prediction method is capable of accurately inferring the vehicle’s attack intention and parameter model, and of effectively reducing the accumulation of prediction errors and the time required for the algorithmic process compared to existing methods.

PGL: A short-time model for ship trajectory prediction

Abstract Ship trajectory prediction plays a critical role in collision detection and risk assessment. To enhance prediction accuracy and efficiency, a novel hybrid particle swarm optimisation (PSO) and grey wolf optimisation (GWO) long, short-term memory (LSTM) network model is proposed (PGL model). The hybrid PSO-GWO optimisation method combines the algorithm's strengths and offers improved stability and performance. The hybrid algorithm is employed to optimise the hyperparameters of the LSTM neural networks to enhance prediction accuracy and efficiency. To demonstrate the superiority of the PGL model, the LSTM, PSO-LSTM and PGL are applied to the same dataset, and then prediction performance and processing time are compared. Experimental results indicate that the proposed PGL algorithm outperforms prediction accuracy and optimisation time.

A real-time approach for surgical activity recognition and prediction based on transformer models in robot-assisted surgery.

This paper presents a deep learning approach to recognize and predict surgical activity in robot-assisted minimally invasive surgery (RAMIS). Our primary objective is to deploy the developed model for implementing a real-time surgical risk monitoring system within the realm of RAMIS. We propose a modified Transformer model with the architecture comprising no positional encoding, 5 fully connected layers, 1 encoder, and 3 decoders. This model is specifically designed to address 3 primary tasks in surgical robotics: gesture recognition, prediction, and end-effector trajectory prediction. Notably, it operates solely on kinematic data obtained from the joints of robotic arm. The model's performance was evaluated on JHU-ISI Gesture and Skill Assessment Working Set dataset, achieving highest accuracy of 94.4% for gesture recognition, 84.82% for gesture prediction, and significantly low distance error of 1.34mm with a prediction of 1s in advance. Notably, the computational time per iteration was minimal recorded at only 4.2ms. The results demonstrated the excellence of our proposed model compared to previous studies highlighting its potential for integration in real-time systems. We firmly believe that our model could significantly elevate realms of surgical activity recognition and prediction within RAS and make a substantial and meaningful contribution to the healthcare sector.

Predicting the evolution from first-episode psychosis to mood or psychotic disorder: A systematic review of biological markers.

The development of paraclinical tools to assist clinical assessment is already widespread in nearly all other medical specialties. In psychiatry, many efforts are being made to improve management strategies using these new techniques. The first episode psychosis (FEP) is a clinical entity whose evolution after onset is difficult to predict in the current state of our practices. Our main objective was to identify from the literature the most promising biological markers for early prediction of thymic or psychotic trajectories following FEP. We performed a systematic literature review on 4 databases: PubMed, Scopus, PsycINFO, and Web of Science following PRISMA guidelines and using search terms related to FEP and biomarkers. Eight studies were included in our final analysis. Several biomarkers showed promising discriminatory capacities for predicting post-FEP evolution: the interleukins IL-6, IL-12p40, IL-1β, and the mRNA expression levels of the DICER-1 and AKT-1 genes. Other studies that opted for broad-spectrum strategies also highlighted new leads for the discovery of additional biomarkers. Overall, our results indicate the value of replicating studies targeting the analysis of the predictive capacities of several biological markers. It also appears important to homogenize methodologies and favor the construction of predictive models on several of these markers to reinforce their statistical significance.

Application of switching-input LSTM network for vessel trajectory prediction

Application of switching-input LSTM network for vessel trajectory prediction

An Explanatory Mixed-methods Study of "ICU net benefit": Triage and Trajectory for Sepsis and Acute Respiratory Failure.

Patients with sepsis and/or acute respiratory failure are at high risk for death or long hospital stays, yet limited evidence exists to guide triage to intensive care units (ICUs) or general medical wards for the majority of these patients who do not initially require life support. To identify factors that influence how hospitals triage patients with capacity-sensitive conditions and those factors that may account for observed ICU relative to ward, or ward relative to ICU, benefits for such patients. We conducted an explanatory sequential mixed-methods study. As part of a 27-hospital, two-health system retrospective cohort study, we calculated hospital-specific measurements of ICU net benefit for patients with sepsis and/or acute respiratory failure. Hospitals among the highest ICU net benefit and lowest ICU net benefit (or highest ward net benefit) from each study health system were selected for in-depth qualitative study. At each hospital interviews were conducted with emergency department (ED), ward, and ICU clinicians and administrators. Interview transcripts were analyzed using flexible coding and the framework method. Interviews were conducted with 118 respondents (46 physicians, 43 nurses, 5 advanced practice providers, and 24 administrators) from four hospitals. Respondents across hospitals agreed that the prediction of patient trajectory is central to triage decisions, but there was variation in opinion across work locations about optimal pre-triage ED interventions in terms of intensity, repetition, clinical reassessment, and observation duration. The main difference observed between high and low ICU net benefit hospitals related to the way respondents working in the ICU and ward described their responses to patients who experience rapid clinical deviations from triage-expected trajectories including sustained lack of critical care needs after admission to the ICU and acute critical care needs after admission to the ward. Hospitals with low ICU net benefit (or high ward net benefit) had particularly robust and proactive rapid response and clinical decompensation surveillance practices for ward-admitted patients. Particularly proactive rapid response programs that deliver on-location critical care may quantitatively increase ward net benefit by bringing ICU benefits without ICU-associated harms to ward patients who become critically ill.

Uncertainty-Aware Multimodal Trajectory Prediction via a Single Inference from a Single Model.

In the domain of autonomous driving, trajectory prediction plays a pivotal role in ensuring the safety and reliability of autonomous systems, especially when navigating complex environments. Unfortunately, trajectory prediction suffers from uncertainty problems due to the randomness inherent in the driving environment, but uncertainty quantification in trajectory prediction is not widely addressed, and most studies rely on deep ensembles methods. This study presents a novel uncertainty-aware multimodal trajectory prediction (UAMTP) model that quantifies aleatoric and epistemic uncertainties through a single forward inference. Our approach employs deterministic single forward pass methods, optimizing computational efficiency while retaining robust prediction accuracy. By decomposing trajectory prediction into velocity and yaw components and quantifying uncertainty in both, the UAMTP model generates multimodal predictions that account for environmental randomness and intention ambiguity. Evaluation on datasets collected by CARLA simulator demonstrates that our model not only outperforms Deep Ensembles-based multimodal trajectory prediction method in terms of accuracy such as minFDE and miss rate metrics but also offers enhanced time to react for collision avoidance scenarios. This research marks a step forward in integrating efficient uncertainty quantification into multimodal trajectory prediction tasks within resource-constrained autonomous driving platforms.

Four-Dimensional Aircraft Trajectory Prediction with a Generative Deep Learning and Clustering Approach

Medium- and long-term four-dimensional (4D) aircraft trajectory prediction (TP) is a critical technology in air traffic management (ATM). This paper addresses the issue of existing medium- and long-term TP methods that are difficult to accurately fit aircraft trajectory data distributions. We propose a 4D TP method based on K-medoids clustering and conditional tabular generative adversarial networks (CTGAN), called C-CTGAN. Comparative experiments with four long short-term memory (LSTM)-based models and the original CTGAN model show that the proposed model’s TP accuracy is significantly higher than others when predicting medium- and long-term trajectories. When using the trajectory datasets without holding and a prediction time span of 10 min, compared to the convolutional neural network (CNN)-LSTM model, the C-CTGAN model reduces the mean absolute errors (MAEs) of core trajectory parameters, such as latitude, longitude, geometric altitude, and ground speed, by 69.89, 15.00, 74.07, and 84.21%, respectively. Compared to the original CTGAN model, the MAE is reduced by 20.43, 39.09, 31.98, and 17.07%, respectively. When using the trajectory datasets with holding, compared to the CNN-LSTM model, the C-CTGAN model shows MAE reductions of 14.08, 23.68, 31.46, and 2.86%, respectively. Compared to the original CTGAN, the reduction is 34.88, 2.69, 23.16, and 73.91%, respectively.

Fully Interactive Graph-Based Trajectory Prediction via Topological Scenario Representation

Fully Interactive Graph-Based Trajectory Prediction via Topological Scenario Representation

WAKE: Towards Robust and Physically Feasible Trajectory Prediction for Autonomous Vehicles With WAvelet and KinEmatics Synergy

WAKE: Towards Robust and Physically Feasible Trajectory Prediction for Autonomous Vehicles With WAvelet and KinEmatics Synergy

Non-destructive degradation pattern decoupling for early battery trajectory prediction via physics-informed learning

The paper proposes a physics-informed model to predict battery lifetime trajectories by computing thermodynamic and kinetic parameters, saving costly data that has not been established for sustainable manufacturing, reuse, and recycling.

Mesoscale Westward Eddy Trajectory Prediction With Memory Augmented Neural Network

Mesoscale Westward Eddy Trajectory Prediction With Memory Augmented Neural Network

Bidirectional Agent-Map Interaction Feature Learning Leveraged by Map-Related Tasks for Trajectory Prediction in Autonomous Driving

Bidirectional Agent-Map Interaction Feature Learning Leveraged by Map-Related Tasks for Trajectory Prediction in Autonomous Driving

DEMO: A Dynamics-Enhanced Learning Model for multi-horizon trajectory prediction in autonomous vehicles

DEMO: A Dynamics-Enhanced Learning Model for multi-horizon trajectory prediction in autonomous vehicles

An AIS Data-Driven Hybrid Approach to Ship Trajectory Prediction

An AIS Data-Driven Hybrid Approach to Ship Trajectory Prediction

IMGCN: interpretable masked graph convolution network for pedestrian trajectory prediction

ABSTRACT Pedestrian trajectory prediction holds significant research value in various fields, such as autonomous driving, autonomous service robots, and human flow monitoring. Two key challenges in pedestrian trajectory prediction are the modeling of pedestrian social interactions and movement factors. Previous methods have not utilized interpretable information to explore complex situations when modeling social interactions. These methods also focus too much on temporal interactions at each moment when modeling movement factors and are therefore susceptible to slight motion changes. To solve the above problems, we propose an Interpretable Masked Graph Convolution Network (IMGCN) for pedestrian trajectory prediction. The IMGCN utilizes interpretable information such as the pedestrian view area, distance, and motion direction to intelligently mask interaction features, resulting in more precise modeling of social interaction and movement factors. Specifically, we design a spatial and a temporal branch to model pedestrians' social interaction and movement factors, respectively. Within the spatial branch, the view-distance mask module masks pedestrian social interaction by determining whether the pedestrian is within a certain distance and view area to achieve more accurate interaction modeling. In the temporal branch, the motion offset mask module masks pedestrian temporal interaction according to the offset degree of their motion direction to achieve accurate modeling of movement factors. Ultimately, the 2D Gaussian distribution parameters of future trajectory points are predicted by the temporal convolution networks for multi-modal trajectory prediction. On the ETH, UCY and SDD datasets, our proposed method outperforms the baseline models in terms of average displacement error and final displacement error. The code is publicly available at https://github.com/Chenwangxing/IMGCN_master.

Assessment and ascertainment in psychiatric molecular genetics: challenges and opportunities for cross-disorder research.

Psychiatric disorders are highly comorbid, heritable, and genetically correlated [1-4]. The primary objective of cross-disorder psychiatric genetics research is to identify and characterize both the shared genetic factors that contribute to convergent disease etiologies and the unique genetic factors that distinguish between disorders [4, 5]. This information can illuminate the biological mechanisms underlying comorbid presentations of psychopathology, improve nosology and prediction of illness risk and trajectories, and aid the development of more effective and targeted interventions. In this review we discuss how estimates of comorbidity and identification of shared genetic loci between disorders can be influenced by how disorders are measured (phenotypic assessment) and the inclusion or exclusion criteria in individual genetic studies (sample ascertainment). Specifically, the depth of measurement, source of diagnosis, and time frame of disease trajectory have major implications for the clinical validity of the assessed phenotypes. Further, biases introduced in the ascertainment of both cases and controls can inflate or reduce estimates of genetic correlations. The impact of these design choices may have important implications for large meta-analyses of cohorts from diverse populations that use different forms of assessment and inclusion criteria, and subsequent cross-disorder analyses thereof. We review how assessment and ascertainment affect genetic findings in both univariate and multivariate analyses and conclude with recommendations for addressing them in future research.

Variable impedance control based on the surgeon’s motion intention for robot-assisted minimally invasive surgery

The paper focuses on incorporating the surgeon’s motion intention into impedance control, which is one of the most commonly used control schemes in master manipulation, to reduce the user’s motion efforts. To achieve this, the non-autoregressive spatio-temporal transformer network is applied to predict the surgeon’s motion trajectory, which is described by the control points of B-spline. The study investigates the effects of input data length, data feature, and feedforward data on trajectory prediction. Furthermore, a Cartesian variable impedance control strategy without force sensor is implemented by adjusting the damping parameters according to the future external force. The experiment shows a promising result, where the interactive force under this control strategy is lower than the one with the current external force, and the precise operation still performs well.

VECTOR: Velocity-Enhanced GRU Neural Network for Real-Time 3D UAV Trajectory Prediction

This paper addresses the challenge of predicting 3D trajectories for Unmanned Aerial Vehicles (UAVs) in real-time, a critical task for applications like aerial surveillance and defense. Current prediction models primarily leverage only position data, which may not provide the most accurate forecasts for UAV movements and usually fail outside the position domain used in the training phase. Our research identifies a gap in utilizing velocity estimates and first-order dynamics to better capture the dynamics and enhance prediction accuracy and generalizability in any position domain. To bridge this gap, we introduce a trajectory prediction scheme using sequence-based neural networks with Gated Recurrent Units (GRUs) to forecast future velocity and positions based on historical velocity estimates instead of position measurements. This approach is designed to improve the predictive capabilities over traditional methods that rely solely on recurrent neural networks (RNNs) or transformers, which can struggle with scalability in this context. Our methodology employs both synthetic and real-world 3D UAV trajectory data, incorporating diverse patterns of agility, curvature, and speed. Synthetic data are generated using the Gazebo robotics simulator and PX4 Autopilot, while real-world data are sourced from the UZH-FPV and Mid-Air drone racing datasets. We train the GRU-based models on drone 3D position and velocity samples to capture the dynamics of UAV movements effectively. Quantitatively, the proposed GRU-based prediction algorithm demonstrates superior performance, achieving a mean square error (MSE) ranging from 2×10−8 to 2×10−7. This performance outstrips existing state-of-the-art RNN models. Overall, our findings confirm the effectiveness of incorporating velocity data in improving the accuracy of UAV trajectory predictions across both synthetic and real-world scenarios, in and out of position data distributions. Finally, we open-source our 5000 trajectories dataset and a ROS2 package to facilitate the integration with existing ROS-based UAV systems.

Efficient multi-target vehicle trajectory prediction based on multi-scale graph convolution

Efficient multi-target vehicle trajectory prediction based on multi-scale graph convolution