Search results for: Hyeongbok Kim

Authors: Hyeongbok Kim, Lingling Zhao, Xiaohong Su, Junjie Wang

Abstract:

The Bernoulli filter is a precise Bayesian filter for single target tracking based on the random finite set theory. The standard Bernoulli filter often underestimates the number of targets. This study proposes a Gaussian particle flow (GPF) Bernoulli filter employing particle flow to migrate particles from prior to posterior positions to improve the performance of the standard Bernoulli filter. By employing the particle flow filter, the computational speed of the Bernoulli filters is significantly improved. In addition, the GPF Bernoulli filter provides a more accurate estimation compared with that of the standard Bernoulli filter. Simulation results confirm the improved tracking performance and computational speed in two- and three-dimensional scenarios compared with other algorithms.

Keywords: Bernoulli filter, particle filter, particle flow filter, random finite sets, target tracking

Procedia PDF Downloads 53

Authors: Hyeongbok Kim, Lingling Zhao, Xiaohong Su

Abstract:

Deep reinforcement learning has been applied to address various problems in robotics, such as autonomous driving and unmanned aerial vehicle. However, because of the sparse reward penalty for a collision with obstacles during the navigation mission, the agent fails to learn the optimal policy or requires a long time for convergence. Therefore, using obstacles and enemy agents, in this paper, we present a curriculum-based boost learning method to effectively train compound skills during multi-agent reinforcement learning. First, to enable the agents to solve challenging tasks, we gradually increased learning difficulties by adjusting reward shaping instead of constructing different learning environments. Then, in a benchmark environment with static obstacles and moving enemy agents, the experimental results showed that the proposed curriculum learning strategy enhanced cooperative navigation and compound collision avoidance skills in uncertain environments while improving learning efficiency.

Keywords: curriculum learning, hard exploration, multi-agent reinforcement learning, robotic navigation, sparse reward

Procedia PDF Downloads 61