About

Ashley Hill’s research sits at the intersection of robotics, reinforcement learning, and state representation learning, with a focus on enabling real-world, off-road mobile robots to navigate safely and efficiently. Her most influential work, “Decoupling feature extraction from policy learning” (24 citations), challenges the end-to-end learning paradigm by demonstrating how state representation learning can dramatically improve sample efficiency when controlling real robots from vision. This foundational insight is complemented by her co-creation of the S-RL Toolbox (20 citations), a widely used open-source benchmark that provides environments, datasets, and evaluation metrics for state representation learning—a resource that has become a standard for researchers in the field. Hill has also pioneered novel approaches for online control gain tuning, using neural networks and the Covariance Matrix Adaptation Evolution Strategy (CMA-ES) to adapt robot behavior in real time under poor grip conditions. Her work on velocity fluctuation and path tracking in off-road settings (8 and 6 citations) directly addresses the practical challenge of balancing speed with safety. Through her development of gradient-based feature importance methods for neural networks, Hill has contributed interpretability tools that help roboticists understand and trust their learned controllers. Her research is characterized by a rare combination of theoretical depth and hands-on robotic implementation, making her a leading voice in bringing reinforcement learning from simulation to rugged, real-world terrains.

Research Focus

Key Achievements

5
H-Index
8
Papers
71
Total Citations
9
Avg Citations/Paper
🏆 Most Cited Paper
Decoupling feature extraction from policy learning: assessing benefits\n of state representation learning in goal based robotics
24 citations · 2019
📈 Most Prolific Year: 2019 (2 Papers)
🤝 Key Collaborators: 9
🏛 Institutions: Laboratoire d’Informatique et Systèmes, Commissariat à l'Énergie Atomique et aux Énergies Alternatives, Laboratoire d'Intégration des Systèmes et des Technologies

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago