Benjamin Evans

About

Benjamin Evans is a rising researcher at the forefront of autonomous systems, specializing in deep reinforcement learning (DRL) and its application to high-speed robotics. His work directly tackles one of the field’s most persistent challenges: bridging the simulation-to-reality gap. In his highly cited 2023 study, "Comparing deep reinforcement learning architectures for autonomous racing" (11 citations), Evans systematically evaluated how different neural network controllers replace traditional perception-planning-control pipelines, providing a critical benchmark for the community. He further advanced the field with his innovative "Bypassing the Simulation-to-Reality Gap" paper (5 citations), where he introduced a supervisor-guided online reinforcement learning method that allows robots to continue learning in the real world, overcoming the limitations of simulated training environments. This approach enables more robust and adaptive control policies for dynamic robotic tasks. By focusing on autonomous racing—a demanding testbed for real-time decision-making—Evans demonstrates how DRL can achieve peak performance under extreme conditions. His work is already shaping how researchers design and deploy learning-based controllers for autonomous vehicles and robots, offering practical solutions to one of robotics’ most stubborn obstacles.

Research Focus

Key Achievements

Top Papers

1. 1
   Comparing deep reinforcement learning architectures for autonomous racing

   11 citations · 2023
2. 2
   Bypassing the Simulation-to-Reality Gap: Online Reinforcement Learning Using a Supervisor

   5 citations · 2023
3. 3
   Comparing Deep Reinforcement Learning Architectures for Autonomous Racing

   2 citations · 2023

Key Collaborators

Contact & Links