Felix Feng

Felix Feng

Research Associate,
Laboratory Manager

Imperial College London

Felix (Yuxiang) Feng is a Research Associate at the Transport Systems and Logistics Laboratory in the Centre for Transport studies. His current research focuses on the the use of automotive system models to facilitate the safety validation of autonomous vehicles.

The primary aim of his work is to develop highly configurable models of automotive systems, that simulate the motions host vehicles and their interactions with other road users.

Before joining TSL, Felix obtained a PhD in the Powertrain and Vehicle Research Centre at the University of Bath. He also received an MSc in Mechatronics from the same institution, and a BEng in Mechatronics from Tongji University.

His doctoral research focused on the development of personalised driver models to investigate the influence of driving style on fuel consumption. His models were calibrated from real world driving data using machine learning techniques. Satisfactory performance was obtained when evaluating the driver models in a procedure anchored to standard WLTC drive cycle test.

Interests
  • Autonomous Vehicles
  • Sensor Fusion
  • Vehicle Dynamics
Education

  • PhD in Mechanical Engineering, 2019

    University of Bath

  • MSc in Mechatronics Engineering, 2014

    University of Bath

  • BE in Mechatronics Engineering, 2013

    Tongji University

Journal Papers

  1. (2022). Monocular Visual Traffic Surveillance: A Review. IEEE Transactions on Intelligent Transportation Systems. Cite DOI
  2. (2022). Spatial-Temporal Flows-Adaptive Street Layout Control Using Reinforcement Learning. Sustainability. Cite DOI

Preprints

  1. Adaptive Road Configurations for Improved Autonomous Vehicle-Pedestrian Interactions using Reinforcement Learning. In review. Preprint. Cite
  2. Location-routing Optimisation for Urban Logistics Using Mobile Parcel Locker Based on Hybrid Q-Learning Algorithm. arXiv, In review. Preprint. PDF Cite DOI
  3. Risk-Aware Stochastic Vehicle Trajectory Prediction with Spatial-Temporal Interaction Modelling. In review. Preprint. Cite
  4. Quantitative Risk Indices for Autonomous Vehicle Training Systems. arXiv. Preprint. PDF Cite DOI

Conference Papers

  1. (2022). Location-routing Optimisation for Urban Logistics Using Mobile Parcel Locker Based on Hybrid Q-Learning Algorithm. Transportation Research Board 101st Annual Meeting (TRBAM 2022). Cite
  2. (2022). Novel Trajectory Prediction Algorithm Using a Full Dataset: Comparison and Ablation Studies. 2022 IEEE 25th International Conference on Intelligent Transportation Systems (ITSC 2022). Cite
  3. (2021). A Reinforcement Learning-based Adaptive Control Model for Future Street Planning: An Algorithm and A Case Study. 2021 57th ISOCARP World Planning Congress (ISOCARP 2021). Cite DOI
  4. (2021). Fast Collision Prediction for Autonomous Vehicles using a Stochastic Dynamics Model. 2021 IEEE 24th International Conference on Intelligent Transportation Systems (ITSC 2021). Cite
  5. (2020). Intelligent Management of On-street Parking Provision for the Autonomous Vehicles Era. 2020 IEEE 23rd International Conference on Intelligent Transportation Systems (ITSC 2020). Cite DOI