Short-term UAV path-planning with monocular-inertial SLAM in the loop

Abstract

Small Unmanned Aerial Vehicles (UAVs) are some of the most promising robotic platforms in a variety of applications due to their high mobility. Their restricted computational and payload capabilities, however, translate into significant challenges in automating their navigation. With Simultaneous Localization And Mapping (SLAM) systems recently demonstrated to be employable onboard UAVs, the focus fall on path-planning on the quest of achieving autonomous navigation. With the vast body of path-planning literature often assuming perfect maps or maps known a priori, the biggest challenge lies in dealing with the robustness and accuracy limitations of onboard SLAM in real missions. In this spirit, this paper proposes a path-planning algorithm designed to work in the loop of the SLAM estimation of a monocular-inertial system. This point-to-point planner is demonstrated to navigate in an unknown environment using the incrementally generated SLAM map, while dictating the navigation strategy for preferable acquisition of sensor data for better estimations within SLAM. A thorough evaluation testbed of both simulated and real data is presented, demonstrating the robustness of the proposed pipeline against the state-of-the-art and its dramatically lower computational complexity, revealing its suitability to UAV navigation.

Perception-aware Receding Horizon Navigation for MAVs

“…Sabdat et al [10] proposed a perception quality metric that combines the number of visible features and the viewing angle with respect to visible surfaces and incorporated this metric into Rapidly-Exploring Random Tree (RRT*) [11] for planning. Alzugaray et al [12] first sampled positions near obstacles based on the intuition that pose estimation error is small when the camera is close to the features on obstacles. Then path planning was carried out based on the sampled positions.…”

Vision Object-Oriented Augmented Sampling-Based Autonomous Navigation for Micro Aerial Vehicles

“…This method cannot make immediate adjustments in the presence of large, irregular obstacles, although it is indispensable in a real environment. The author proposes that the trajectory follows obstacles and always maintains a safety range [5] when MAVs face complex environments. Although this method guarantees sufficient safety, planning is restricted.…”

ME‐MADDPG: An efficient learning‐based motion planning method for multiple agents in complex environments

“…Some direct replanned policies must be generated as soon as possible so that all the agents can adapt to the changing environment 10 . It becomes intractable for some search‐based approaches 11–16 because continuous online replanning requires high timeliness that many search‐based algorithms could not reach.…”