SEER: Safe Efficient Exploration for Aerial Robots using Learning to Predict Information Gain @KumarLabPenn
SEER: Safe Efficient Exploration for Aerial Robots using Learning to Predict Information Gain  @KumarLabPenn
Uploaded September 2022 | Updated November 2026, 5 hours ago
We address the problem of efficient 3-D exploration in indoor environments for micro aerial vehicles with limited sensing capabilities and payload/power constraints. We develop an indoor exploration framework that uses learning to predict the occupancy of unseen areas, extracts semantic features, samples viewpoints to predict information gains for different exploration goals, and plans informative trajectories to enable safe and smart exploration. Extensive experimentation in simulated and real-world environments shows the proposed approach outperforms the state-of-the-art exploration framework by 24% in terms of the total path length in a structured indoor environment and with a higher success rate during exploration.

arxiv: arxiv.org/abs/2209.11034
SEER: Safe Efficient Exploration for Aerial Robots using Learning to Predict Information GainAggressive Flight 2017Robust Fruit Counting: Combining Deep Learning, Tracking, and Structure from MotionMicro Assembly Using Magnetic RobotsExperiments in Fast, Autonomous, GPS-Denied Quadrotor FlightToward Image Based Visual Servoing for Aerial Grasping and PerchingNetworks of Leaders and FollowersTightly-Coupled Monocular Visual-Inertial Fusion for Autonomous Flight of Rotorcraft MAVsEvMAPPER: High Altitude Orthomapping with Event CamerasRobust Aerial Robot Swarms Without Collision AvoidancePennovationCooperation in Construction
Vijay Kumar |

SEER: Safe Efficient Exploration for Aerial Robots using Learning to Predict Information Gain

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER