Search Results for author: Zhiyu Huang

Found 5 papers, 2 papers with code

Augmenting Reinforcement Learning with Transformer-based Scene Representation Learning for Decision-making of Autonomous Driving

no code implementations24 Aug 2022 Haochen Liu, Zhiyu Huang, Xiaoyu Mo, Chen Lv

Decision-making for urban autonomous driving is challenging due to the stochastic nature of interactive traffic participants and the complexity of road structures.

Autonomous Driving Decision Making +3

STrajNet: Multi-modal Hierarchical Transformer for Occupancy Flow Field Prediction in Autonomous Driving

1 code implementation31 Jul 2022 Haochen Liu, Zhiyu Huang, Chen Lv

Therefore, this paper proposes a novel Multi-modal Hierarchical Transformer network that fuses the vectorized (agent motion) and visual (scene flow, map, and occupancy) modalities and jointly predicts the flow and occupancy of the scene.

Autonomous Driving

Multi-Agent Trajectory Prediction With Heterogeneous Edge-Enhanced Graph Attention Network

1 code implementation IEEE Transactions on Intelligent Transportation Systems 2022 Xiaoyu Mo, Zhiyu Huang, Yang Xing, Chen Lv

Simultaneous trajectory prediction for multiple heterogeneous traffic participants is essential for safe and efficient operation of connected automated vehicles under complex driving situations.

Autonomous Vehicles Decision Making +3

Prioritized Experience-based Reinforcement Learning with Human Guidance: Methdology and Application to Autonomous Driving

no code implementations26 Sep 2021 Jingda Wu, Zhiyu Huang, Wenhui Huang, Chen Lv

A novel prioritized experience replay mechanism that adapts to human guidance in the reinforcement learning process is proposed to boost the efficiency and performance of the reinforcement learning algorithm.

Autonomous Driving online learning +1

Uncertainty-Aware Model-Based Reinforcement Learning with Application to Autonomous Driving

no code implementations23 Jun 2021 Jingda Wu, Zhiyu Huang, Chen Lv

Then, a novel uncertainty-aware model-based RL framework is developed based on the adaptive truncation approach, providing virtual interactions between the agent and environment model, and improving RL's training efficiency and performance.

Autonomous Driving Model-based Reinforcement Learning +1

Cannot find the paper you are looking for? You can Submit a new open access paper.