Deep reinforcement learning for drone navigation using sensor data

Abstract

Mobile robots such as unmanned aerial vehicles (drones) can be used for surveillance, monitoring and data collection in buildings, infrastructure and environments. The importance of accurate and multifaceted monitoring is well known to identify problems early and prevent them escalating. This motivates the need for flexible, autonomous and powerful decision-making mobile robots. These systems need to be able to learn through fusing data from multiple sources. Until very recently, they have been task specific. In this paper, we describe a generic navigation algorithm that uses data from sensors on-board the drone to guide the drone to the site of the problem. In hazardous and safety-critical situations, locating problems accurately and rapidly is vital. We use the proximal policy optimisation deep reinforcement learning algorithm coupled with incremental curriculum learning and long short-term memory neural networks to implement our generic and adaptable navigation algorithm. We evaluate different configurations against a heuristic technique to demonstrate its accuracy and efficiency. Finally, we consider how safety of the drone could be assured by assessing how safely the drone would perform using our navigation algorithm in real-world scenarios.

Robust Data Fusion of UAV Navigation Measurements with Application to the Landing System

“…So we reduce our task to (10). In [40] it is recommended to use the proposed algorithm on personal computer.…”

“…In recent years the use of machine learning methods has become traditional and these methods show good performance in 3D analysis and reconstruction of the earth surface [6,7]. These methods are applicable to UAV navigation in unknown environment [8][9][10]. However, they are hardly applicable in the navigation and control of objects with continuous and noisy dynamics.…”

Deep reinforcement learning based mobile robot navigation: A review

“…(1) Expansion of network input Considering that the agent cannot uniquely distinguish its state based on current observations, the simplest solution is to add several previous observation frames as network inputs to improve its ability to distinguish among states [36,51,72,75,76] . In addition, previous rewards and actions also contain state information, so some studies have input previous rewards and actions to the network [33,44,63,77] . Another input expansion technique is the two-stream Q network proposed by Wang et al [78] , which adds the difference between two frames of laser scanning data.…”

A Reinforcement Learning Routing Protocol for UAV Aided Public Safety Networks

“…In [49], authors used drones for surveillance and data collection in buildings. Sensors are used by the drones to navigate the buildings to identify and pinpoint the problems.…”