KHAOS: a Kinematic Human Aware Optimization-based System for Reactive Planning of Flying-Coworker

Abstract

The use of drones in human-populated areas is increasing day by day. Such robots flying in close proximity to humans and potentially interacting with them, as in object handover or delivery, need to carefully plan their navigation considering the presence of humans. We propose a humanaware 3D reactive planner based on stochastic optimization for drone navigation. Besides considering the kinematics constraints of the drone, we propose two criteria to produce socially acceptable trajectories. The first, called discomfort, considers the unease caused to the humans spatially close to fast-moving drones. The second, called visibility, promotes the drone's visibility for humans. We demonstrate the planner's performance and adaptability in various simulated experiments.

Bridging Requirements, Planning, and Evaluation: A Review of Social Robot Navigation

“…Similarly, an optimal control algorithm, but for sustaining formations of various structures, was proposed in [331]. On the other hand, Truc et al [332] developed a 3D reactive planner for human-aware drone navigation in populated environments that is based on a stochastic optimization of discomfort caused by the drone's proximity to pedestrians and the visibility of the drone.…”

Reactive Planning for Coordinated Handover of an Autonomous Aerial Manipulator

“…(1) where C closest human hand is the distance from the grid cell to the nearest human hand, C vis is the visibility cost defined in [Truc et al(2022)] considering human's visual field and the effort to turn the head to see the robot, C turn is an angle originating in the human torso and formed between the front of the human torso and the cell and C torso dist is the distance from the cell to the human torso.…”

“…where C vis is the visibility cost defined in [Truc et al(2022)] considering human's visual field and the effort to turn the head to see the robot and C human head dist is the distance between the end of the shared object caught by the AAM's end effector and the human's head base goal grid From the list of configurations of the shared object, we now know the possible positions for the end effector . Starting from the position where C shared object is minimal, we generate a new cost grid named base goal grid which this time allows us to determine the position of the AAM base.…”

“…To deal with the case of coordinated motion for an AAM performing a handover task, we propose an extension of KHAOS presented in [Truc et al(2022)]. Complete state representation of the AAM is given as follows:…”

“…They have all the necessary components to move and interact in a human-populated environment. KHAOS presented in [Truc et al(2022)] provides a solution to generate human-aware trajectories for social navigation of aerial robot. These trajectories may allow an AAM to navigate in an environment populated with humans and various dynamic obstacles.…”