Trajectory Distribution

Trajectory distribution research focuses on modeling and predicting the probability of different possible paths or sequences of states, particularly in complex systems like robot motion planning and human behavior prediction. Current research emphasizes learning these distributions using advanced models such as diffusion models and normalizing flows, often incorporating techniques like posterior sampling and trajectory weighting to improve robustness and efficiency. This work has significant implications for various fields, including robotics (safe and efficient motion planning), autonomous driving (predicting and reacting to other agents), and human-computer interaction (more realistic and adaptable AI systems).

Papers

November 15, 2024

BMP: Bridging the Gap between B-Spline and Movement Primitives
Weiran Liao, Ge Li, Hongyi Zhou, Rudolf Lioutikov, Gerhard Neumann
Imitation Learning Motion Primitive B Spline Motion Representation Trajectory Distribution Rasterized Image

March 30, 2024

Joint Pedestrian Trajectory Prediction through Posterior Sampling
Haotian Lin, Yixiao Wang, Mingxiao Huo, Chensheng Peng, Zhiyuan Liu, Masayoshi Tomizuka
Trajectory Prediction Conditional Diffusion Model Posterior Sampling Multi View Diffuser Trajectory Distribution

August 3, 2023

Motion Planning Diffusion: Learning and Planning of Robot Motions with Diffusion Models
Joao Carvalho, An T. Le, Mark Baierl, Dorothea Koert, Jan Peters
LeArning Abstract Diffusion Model Task Planning Robot Motion Diffusion Planner Trajectory Distribution Learned Prior

June 22, 2023

Harnessing Mixed Offline Reinforcement Learning Datasets via Trajectory Weighting
Zhang-Wei Hong, Pulkit Agrawal, Rémi Tachet des Combes, Romain Laroche
Offline Reinforcement Learning Offline Reinforcement Learning Algorithm Expert Trajectory Trajectory Distribution

June 1, 2023

Safe Offline Reinforcement Learning with Real-Time Budget Constraints
Qian Lin, Bo Tang, Zifan Wu, Chao Yu, Shangqin Mao, Qianlong Xie, Xingxing Wang, Dong Wang
Reinforcement Learning Budget Constraint Offline Safe Reinforcement Learning Trajectory Distribution Multi Session Budget Optimization

April 11, 2023

Learning Distributions over Trajectories for Human Behavior Prediction
Anna Mészáros, Julian F. Schumann, Javier Alonso-Mora, Arkady Zgonnikov, Jens Kober
Handwritten Trajectory Future Trajectory Behavior Prediction Generative Trajectory Trajectory Distribution

November 20, 2022

The Past Does Matter: Correlation of Subsequent States in Trajectory Predictions of Gaussian Process Models
Steffen Ridderbusch, Sina Ober-Blöbaum, Paul Goulart
Gaussian Process Trajectory Prediction State Space Model Total Correlation Uncertainty Propagation Trajectory Distribution Structured State Evolution

November 3, 2022

Safe Real-World Autonomous Driving by Learning to Predict and Plan with a Mixture of Experts
Stefano Pini, Christian S. Perone, Aayush Ahuja, Ana Sofia Rufino Ferreira, Moritz Niendorf, Sergey Zagoruyko
Autonomous Vehicle Mixture Component Expert Knowledge Trajectory Generation Vehicle Trajectory High Level Plan Trajectory Distribution

October 26, 2022

Planning with Occluded Traffic Agents using Bi-Level Variational Occlusion Models
Filippos Christianos, Peter Karkus, Boris Ivanovic, Stefano V. Albrecht, Marco Pavone
Autonomous Vehicle Task Planning Agent Trajectory Traffic Agent Trajectory Distribution

September 13, 2022

A Gaussian variational inference approach to motion planning
Hongzhe Yu, Yongxin Chen
Motion Planning Motion Planning Problem Motion Planning Algorithm Trajectory Distribution Gaussian Variational Inference

April 19, 2022

Interventional Behavior Prediction: Avoiding Overly Confident Anticipation in Interactive Prediction
Chen Tang, Wei Zhan, Masayoshi Tomizuka
High Confidence Behavior Prediction Trajectory Distribution Interactive Prediction Conditional Behavior Prediction

January 29, 2022

Explaining Reinforcement Learning Policies through Counterfactual Trajectories
Julius Frost, Olivia Watkins, Eric Weiner, Pieter Abbeel, Trevor Darrell, Bryan Plummer, Kate Saenko
Reinforcement Learning RL Agent Agent Task Trajectory Distribution Multi Agent Rollout

Trajectory Distribution

Papers

BMP: Bridging the Gap between B-Spline and Movement Primitives

Joint Pedestrian Trajectory Prediction through Posterior Sampling

Motion Planning Diffusion: Learning and Planning of Robot Motions with Diffusion Models

Harnessing Mixed Offline Reinforcement Learning Datasets via Trajectory Weighting

Safe Offline Reinforcement Learning with Real-Time Budget Constraints

Learning Distributions over Trajectories for Human Behavior Prediction

The Past Does Matter: Correlation of Subsequent States in Trajectory Predictions of Gaussian Process Models

Safe Real-World Autonomous Driving by Learning to Predict and Plan with a Mixture of Experts

Planning with Occluded Traffic Agents using Bi-Level Variational Occlusion Models

A Gaussian variational inference approach to motion planning

Interventional Behavior Prediction: Avoiding Overly Confident Anticipation in Interactive Prediction

Explaining Reinforcement Learning Policies through Counterfactual Trajectories