GAMDTP: Dynamic Trajectory Prediction with Graph Attention Mamba Network
On the map: State-space sequence models (S4, Mamba) → Video, motion and generative Mamba models
Abstract (arXiv)
Accurate motion prediction of traffic agents is crucial for the safety and stability of autonomous driving systems. In this paper, we introduce GAMDTP, a novel graph attention-based network tailored for dynamic trajectory prediction. Specifically, we fuse the result of self attention and mamba-ssm through a gate mechanism, leveraging the strengths of both to extract features more efficiently and accurately, in each graph convolution layer. GAMDTP encodes the high-definition map(HD map) data and the agents' historical trajectory coordinates and decodes the network's output to generate the final prediction results. Additionally, recent approaches predominantly focus on dynamically fusing historical forecast results and rely on two-stage frameworks including proposal and refinement. To further enhance the performance of the two-stage frameworks we also design a scoring mechanism to evaluate the prediction quality during the proposal and refinement processes. Experiments on the Argoverse dataset demonstrates that GAMDTP achieves state-of-the-art performance, achieving superior accuracy in dynamic trajectory prediction.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in Video, motion and generative Mamba models, one of 10 directions in State-space sequence models (S4, Mamba). Builds on 37 of its 53 references that are in the map. First posted on arXiv 7 Apr 2025.
Builds on
37 of 53 references.
Closest ancestors · 2
Same direction, newest first
- 2024Motion Mamba: Efficient and Long Sequence Motion Generation12 Mar
- 2024VideoMamba: State Space Model for Efficient Video Understanding11 Mar
Other references · 35
- 2017Attention Is All You Needcited by 60,361
- 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,639
- 2017Graph Attention Networkscited by 8,923
- 2020End-to-End Object Detection with Transformerscited by 7,414
- 2023Mamba: Linear-Time Sequence Modeling with Selective State Spacescited by 4,718
- 2021Efficiently Modeling Long Sequences with Structured State Spacescited by 2,465
- 2024Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Dualitycited by 1,336
- 2024Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Modelcited by 1,056
- 2019Argoverse: 3D Tracking and Forecasting with Rich Mapscited by 732
- 2020VectorNet: Encoding HD Maps and Agent Dynamics from Vectorized Representationcited by 530
- 2019MultiPath: Multiple Probabilistic Anchor Trajectory Hypotheses for Behavior Predictioncited by 420
- 2024Jamba: A Hybrid Transformer-Mamba Language Modelcited by 401
23 more
- 2020Learning Lane Graph Representations for Motion Forecastingcited by 382
- 2017DESIRE: Distant Future Prediction in Dynamic Scenes with Interacting Agentscited by 310
- 2021DenseTNT: End-to-end Trajectory Prediction from Dense Goal Setscited by 295
- 2019CoverNet: Multimodal Behavior Prediction using Trajectory Setscited by 259
- 2021MultiPath++: Efficient Information Fusion and Trajectory Aggregation for Behavior Predictioncited by 246
- 2019INTERACTION Dataset: An INTERnational, Adversarial and Cooperative moTION Dataset in Interactive Driving Scenarios with Semantic Mapscited by 231
- 2018Convolutional Social Pooling for Vehicle Trajectory Predictioncited by 221
- 2021Multimodal Motion Prediction with Stacked Transformerscited by 208
- 2021Scene Transformer: A unified architecture for predicting multiple agent trajectoriescited by 165
- 2021HOME: Heatmap Output for future Motion Estimationcited by 122
- 2022HDGT: Heterogeneous Driving Graph Transformer for Multi-Agent Trajectory Prediction via Scene Encodingcited by 110
- 2021THOMAS: Trajectory Heatmap Output with learned Multi-Agent Samplingcited by 101
- 2021Multi-modal Motion Prediction with Transformer-based Neural Network for Autonomous Drivingcited by 73
- 2019Recurrent Neural Networks (RNNs): A gentle Introduction and Overviewcited by 68
- 2024Jamba-1.5: Hybrid Transformer-Mamba Models at Scalecited by 50
- 2023Traj-MAE: Masked Autoencoders for Trajectory Predictioncited by 42
- 2022FJMP: Factorized Joint Multi-Agent Motion Prediction over Learned Directed Acyclic Interaction Graphscited by 41
- 2024HPNet: Dynamic Trajectory Forecasting with Historical Prediction Attentioncited by 38
- 2024SmartRefine: A Scenario-Adaptive Refinement Framework for Efficient Motion Predictioncited by 34
- 2018Naturalistic Driver Intention and Path Prediction using Recurrent Neural Networkscited by 29
- 2024DenseMamba: State Space Models with Dense Hidden Connection for Efficient Large Language Modelscited by 22
- 2023End-to-End Heterogeneous Graph Neural Networks for Traffic Assignmentcited by 13
- 2024Combining Graph Neural Network and Mamba to Capture Local and Global Tissue Spatial Relationships in Whole Slide Imagescited by 7
Built on by
Citing papers per year: 2025: 2.
2 papers.
Similar, not cited
Close in wording, no citation either way
- 2025Trajectory Mamba: Efficient Attention-Mamba Forecasting Model Based on Selective SSM
- 2026GAMMA-Net: Adaptive Long-Horizon Traffic Spatio-Temporal Forecasting Model based on Interleaved Graph Attention and Multi-Axis Mamba
- 2026MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- 2022Wayformer: Motion Forecasting via Simple & Efficient Attention Networks
- 2025Generalizable Trajectory Prediction via Inverse Reinforcement Learning with Mamba-Graph Architecture
- 2024Map-Free Trajectory Prediction with Map Distillation and Hierarchical Encoding
- 2023Utilizing Hybrid Trajectory Prediction Models to Recognize Highly Interactive Traffic Scenarios ↗
- 2025HAMF: A Hybrid Attention-Mamba Framework for Joint Scene Context Understanding and Future Motion Representation Learning