Mario Ynocente Castro
Papers
2
Total Citations
9
H-Index
2
About
Mario Ynocente Castro is a researcher advancing the safety and autonomy of robotic systems, with a focus on autonomous vehicles and mobile manipulation. His work addresses critical challenges in deploying robots in unstructured environments, particularly through simulation-based testing and reinforcement learning. In his highly cited 2020 paper, "Discovering Avoidable Planner Failures of Autonomous Vehicles using Counterfactual Analysis in Behaviorally Diverse Simulation" (6 citations), Castro introduced a novel framework for testing autonomous vehicle planners. By using counterfactual analysis and behaviorally diverse simulations, his method systematically uncovers safety-critical planner failures before real-world deployment, directly improving the reliability of decision-making in automated vehicles. Additionally, in "Distributed Reinforcement Learning of Targeted Grasping with Active Vision for Mobile Manipulators" (3 citations), he presented one of the first RL-based systems for mobile manipulators, enabling robots to perform targeted grasping in unstructured settings. This work integrates active vision and distributed learning, pushing the boundaries of personal robotics. Castro’s contributions are pivotal for ensuring both the safety of autonomous systems and the versatility of robotic manipulation, making him a notable figure in the intersection of robotics, simulation, and reinforcement learning.
Research Focus
Key Achievements
Top Papers
- 1
- 2