• Corpus ID: 226221926

Multi-agent Trajectory Prediction with Fuzzy Query Attention

  title={Multi-agent Trajectory Prediction with Fuzzy Query Attention},
  author={Nitin Kamra and Hao Zhu and Dweep Trivedi and Ming Zhang and Yan Liu},
Trajectory prediction for scenes with multiple agents and entities is a challenging problem in numerous domains such as traffic prediction, pedestrian tracking and path planning. We present a general architecture to address this challenge which models the crucial inductive biases of motion, namely, inertia, relative motion, intents and interactions. Specifically, we propose a relational model to flexibly model interactions between agents in diverse environments. Since it is well-known that… 

Figures and Tables from this paper

M2I: From Factored Marginal Trajectory Prediction to Interactive Prediction
This work exploits the underlying relations between interacting agents and decouple the joint prediction problem into marginal prediction problems, and leverages a marginal prediction model and a conditional prediction model to predict trajectories for the influencers and reactors, respectively.
LatentFormer: Multi-Agent Transformer-Based Interaction Modeling and Trajectory Prediction
LatentFormer, a transformerbased model for predicting future vehicle trajectories, is proposed that leverages a novel technique for modeling interactions among dynamic objects in the scene and achieves state-of-the-art performance and improves upon trajectory metrics by up to 40%.
Tra2Tra: Trajectory-to-Trajectory Prediction With a Global Social Spatial-Temporal Attentive Neural Network
A novel Global Social Spatial-Temporal Attentive Neural Network for trajectory-to-trajectory prediction (Tra2Tra), which can achieve better performance than the state-of-the-art methods not only on two pedestrian-walking datasets, but also on three other complex trajectory datasets, i.e. Collisions, NGsim and Charges.
GRIN: Generative Relation and Intention Network for Multi-agent Trajectory Prediction
A conditional deep generative model that combines advances in graph neural networks and prior and recognition to predict multi-modal future trajectory distribution and it is shown that this model outperforms previous work in multiple performance metrics.
On Adversarial Robustness of Trajectory Prediction for Autonomous Vehicles
This work studies the adversarial robustness of trajectory prediction models by proposing a new adversarial attack that perturbs normal vehicle trajectories to maximize the prediction error and explores possible mitigation techniques via data augmentation and trajectory smoothing.
Self-Growing Spatial Graph Network for Context-Aware Pedestrian Trajectory Prediction
  • Sirin Haddad, S. Lam
  • Computer Science
    2021 IEEE International Conference on Image Processing (ICIP)
  • 2021
This work proposes Social Trajectory Recommender-Gated Graph Recurrent Neighborhood Network (STR-GGRNN), which uses data-driven adaptive online neighborhood recommendation based on the contextual scene features and pedestrian visual cues to predict pedestrians’ trajectories.
Model-based Multi-agent Policy Optimization with Adaptive Opponent-wise Rollouts
A novel decentralized model-based MARL method, named Adaptive Opponent-wise Rollout Policy Optimization (AORPO), is proposed, which can achieve improved sample efficiency with comparable asymptotic performance over the compared MARL methods.


TrafficPredict: Trajectory Prediction for Heterogeneous Traffic-Agents
The proposed long short-term memory-based (LSTM-based) realtime traffic prediction algorithm, TrafficPredict, uses an instance layer to learn instances' movements and interactions and has a category layer to learning the similarities of instances belonging to the same type to refine the prediction.
Social LSTM: Human Trajectory Prediction in Crowded Spaces
This work proposes an LSTM model which can learn general human movement and predict their future trajectories and outperforms state-of-the-art methods on some of these datasets.
DESIRE: Distant Future Prediction in Dynamic Scenes with Interacting Agents
The proposed Deep Stochastic IOC RNN Encoder-decoder framework, DESIRE, for the task of future predictions of multiple interacting agents in dynamic scenes significantly improves the prediction accuracy compared to other baseline methods.
Multi-agent Trajectory Prediction and Generation with Topological Invariants Enforced by Hamiltonian Dynamics
This framework is based on a novel technique for computationally efficient multi-agent trajectory generation from symbolic topological specifications that allows an agent to generate a diverse set of potential future scene evolutions in the form of Cartesian, multi- agent trajectory representations.
Social Attention: Modeling Attention in Human Crowds
This work proposes Social Attention, a novel trajectory prediction model that captures the relative importance of each person when navigating in the crowd, irrespective of their proximity, and demonstrates the performance against a state-of-the-art approach on two publicly available crowd datasets.
You'll never walk alone: Modeling social behavior for multi-target tracking
A model of dynamic social behavior, inspired by models developed for crowd simulation, is introduced, trained with videos recorded from birds-eye view at busy locations, and applied as a motion model for multi-people tracking from a vehicle-mounted camera.
An Evaluation of Trajectory Prediction Approaches and Notes on the TrajNet Benchmark
It is shown that a Recurrent-Encoder with a Dense layer stacked on top, referred to as RED-predictor, is able to achieve sophisticated results compared to elaborated models in such scenarios and some recommendations for overcoming demonstrated shortcomings are given.
Generating Long-term Trajectories Using Deep Hierarchical Networks
This work proposes a hierarchical policy class that automatically reasons about both long-term and short-term goals, which is instantiate as a hierarchical neural network and generates significantly more realistic trajectories compared to non-hierarchical baselines as judged by professional sports analysts.
Generating Multi-Agent Trajectories using Programmatic Weak Supervision
This work presents a hierarchical framework that can effectively learn sequential generative models for capturing coordinated multi-agent trajectory behavior, such as offensive basketball gameplay, and is inspired by recent work on leveraging programmatically produced weak labels, which it extends to the spatiotemporal regime.