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/

Zendy Plus

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

Zendy Tools

Zendy Open

Reinforcement learning (RL) is a promising approach for robotic navigation,allowing robots to learn through trial and error. However, real-world robotictasks often suffer from sparse rewards, leading to inefficient exploration andsuboptimal policies due to sample inefficiency of RL. In this work, weintroduce Confidence-Controlled Exploration (CCE), a novel method that improvessample efficiency in RL-based robotic navigation without modifying the rewardfunction. Unlike existing approaches, such as entropy regularization and rewardshaping, which can introduce instability by altering rewards, CCE dynamicallyadjusts trajectory length based on policy entropy. Specifically, it shortenstrajectories when uncertainty is high to enhance exploration and extends themwhen confidence is high to prioritize exploitation. CCE is a principled andpractical solution inspired by a theoretical connection between policy entropyand gradient estimation. It integrates seamlessly with on-policy and off-policyRL methods and requires minimal modifications. We validate CCE acrossREINFORCE, PPO, and SAC in both simulated and real-world navigation tasks. CCEoutperforms fixed-trajectory and entropy-regularized baselines, achieving an18\% higher success rate, 20-38\% shorter paths, and 9.32\% lower elevationcosts under a fixed training sample budget. Finally, we deploy CCE on aClearpath Husky robot, demonstrating its effectiveness in complex outdoorenvironments.

Confidence-Controlled Exploration: Efficient Sparse-Reward Policy Learning for Robot Navigation