English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Deep Reinforcement Learning is quickly becoming a popular method for trainingautonomous Unmanned Aerial Vehicles (UAVs). Our work analyzes the effects ofmeasurement uncertainty on the performance of Deep Reinforcement Learning (DRL)based waypoint navigation and obstacle avoidance for UAVs. Measurementuncertainty originates from noise in the sensors used for localization anddetecting obstacles. Measurement uncertainty/noise is considered to follow aGaussian probability distribution with unknown non-zero mean and variance. Weevaluate the performance of a DRL agent trained using the Proximal PolicyOptimization (PPO) algorithm in an environment with continuous state and actionspaces. The environment is randomized with different numbers of obstacles foreach simulation episode in the presence of varying degrees of noise, to capturethe effects of realistic sensor measurements. Denoising techniques like the lowpass filter and Kalman filter improve performance in the presence of unbiasednoise. Moreover, we show that artificially injecting noise into themeasurements during evaluation actually improves performance in certainscenarios. Extensive training and testing of the DRL agent under various UAVnavigation scenarios are performed in the PyBullet physics simulator. Toevaluate the practical validity of our method, we port the policy trained insimulation onto a real UAV without any further modifications and verify theresults in a real-world environment.

Sim-to-Real Deep Reinforcement Learning based Obstacle Avoidance for UAVs under Measurement Uncertainty