Indoor GPS-denied Context Based SLAM Aided Guidance for Autonomous Unmanned Aerial Systems

Abstract

Autonomous exploration and mapping of environments is an important problem in robotics. Efficient exploration of structured environments requires that the robot utilize region-specific exploration strategies and coordinate with search other agents. This paper details the exploration and guidance system of a multi-quadrotor unmanned aerial system (UAS) capable of exploring cluttered indoor areas without relying on any external aides. Specifically, a graph-based frontier search algorithm which is aided by an onboard Simultaneous Localization and Mapping (SLAM) system is developed and flight tested. A technique is developed in for segmenting an indoor office-like environment into regions and to utilize the SLAM map to conduct specific activities in these regions. A goal-directed exploration strategy is created building on existing hybrid deliberative-reactive approaches to exploration. An obstacle avoidance and guidance system is implemented to ensure that the vehicle explores maximum indoor area while avoiding obstacles. The environment is explored and regions are segmented by detecting rooms and hallways which expedites the search. The multi-vehicle system is Georgia Tech Aerial Robotic Team's entry for the annual International Aerial Robotics Competition (IARC).

UAV Control in Autonomous Object-Goal Navigation: A Systematic Literature Review

“…Ten papers focus on the planning method as a graph-based approach, using a cost function over the graph's vertices to determine which point to reach. Among the graph-based methods, tree-search algorithms were found, such as Rapidly-exploring Random Tree (RRT) (Dang et al, 2018;Prophet and Trommer, 2020), its improved version RRT* (Alarcon et al, 2021), Dijkstra algorithm (Bershadsky and Johnson, 2013), and frontierbased exploration (Bachrach et al, 2010). The main idea behind graph-based methods is to go from one node to another, traveling the shortest obstaclefree path through vertex cost analysis between the required nodes.…”

“…As shown in Table 10, nine papers treat the mapping task as reconstruction, fusing sensory data to create a spatial projection for each element in the scene. Simultaneous mapping and localization (SLAM) (Bachrach et al, 2010;Bershadsky and Johnson, 2013;Dang et al, 2020;Zhao et al, 2021b), a widely used method in navigation, solves this reconstruction problem. SLAM fuses internal and external sensor information to create a map with structural information such as corridors, objects, and the vehicle itself.…”

“…One category was named probabilistic, with seventeen articles using this kind of method, processing the vehicle sensor data with probability techniques to infer an expected state. In this category was included the SLAM technique (Bachrach et al, 2010;Bershadsky and Johnson, 2013;Cocchioni et al, 2014;Dang et al, 2020;Narazaki et al, 2022;Kumar et al, 2020;Zhao et al, 2021b) because the sensor data is processed to correct the current vehicle pose. Additionally, SLAM is complemented with probability methods such as Kalman filters with its variations of extended Kalman filters (EKF) (Konrad et al, 2017;LI and HU, 2021;Mansouri et al, 2020b;Meng et al, 2019) and Kalman consensus filter (Li et al, 2018), non-linear observer (Helgesen et al, 2019), particle filters (Wang et al, 2017), and Monte Carlo processes (Delamer et al, 2021;Kusnur et al, 2021).…”

“…In exploration and navigation behavior, the control task relies on sensory information to compute its internal state to achieve momentum. This internal state describes the current vehicle control condition regarding information such Bershadsky and Johnson (2013); Nakamura et al (2016) as position, pose, and heading angle. This internal state's accurate description and estimation improve the control task's performance, allowing a safe and smooth movement.…”

“…Sonar technology is commonly used in maritime vehicles such as submarines. Bershadsky and Johnson (2013); Kumar et al (2020), combines the LiDAR and Sonar sensors to achieve autonomous navigation. Only You et al (2020) uses Radio detection and ranging (Radar), using radio waves to determine the distance, angle, and radial velocity of objects.…”