Comparison of Trajectory Optimization Algorithms for High-Speed Quadrotor Flight Near Obstacles

Morrell, Benjamin; Thakker, Rohan; Merewether, Gene; Reid, Robert; Rigter, Marc; Tzanetos, Theodore; Chamitoff, Gregory

doi:10.1109/lra.2018.2868866

Cited by 21 publications

(11 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Morrell et al (2018) also propose a similar method, with the additional options to constrain the optimization within computed free‐space corridors in the space or respect other hard constraints in the optimization. They propose multiple variants, ranging from conservatively planning only through convex regions of free space to treating unknown space as free.…”

Section: Local Replanningmentioning

confidence: 99%

An open‐source system for vision‐based micro‐aerial vehicle mapping, planning, and flight in cluttered environments

Oleynikova

Lanegger

Taylor

et al. 2020

Journal of Field Robotics

View full text Add to dashboard Cite

We present an open-source system for Micro-Aerial Vehicle (MAV) autonomous navigation from vision-based sensing. Our system focuses on dense mapping, safe local planning, and global trajectory generation, especially when using narrow fieldof-view sensors in very cluttered environments. In addition, details about other necessary parts of the system and special considerations for applications in realworld scenarios are presented. We focus our experiments on evaluating global planning, path smoothing, and local planning methods on real maps made on MAVs in realistic search-and-rescue and industrial inspection scenarios. We also perform thousands of simulations in cluttered synthetic environments, and finally validate the complete system in real-world experiments. K E Y W O R D S aerial robotics, GPS-denied operation, mapping, obstacle avoidance, planning 1 | INTRODUCTION Autonomous navigation from on-board sensing is essential for Micro-Aerial Vehicles (MAVs) in many applications. In this study, we specifically target applications where MAVs can assist human operators in difficult tasks, such as search-and-rescue (S&R) and industrial inspection appli-Helen Oleynikova: work performed while at ETH, now with Microsoft. Zachary Taylor: work performed while at ETH, now with Oculus.

show abstract

Section: Local Replanningmentioning

confidence: 99%

An open‐source system for vision‐based micro‐aerial vehicle mapping, planning, and flight in cluttered environments

Oleynikova

Lanegger

Taylor

et al. 2020

Journal of Field Robotics

View full text Add to dashboard Cite

show abstract

“…Due to the generalization ability of neural networks [26], the resulting policy performs local collision avoidance, while following a global reference path. To plan trajectories in a limited time period, that are safe, collision free, and dynamically feasible, is achieved by [27]. In [28], the maximum latency for the robot to guarantee safety is related to the desired speed, the range of its sensing pipeline, and the actuation limitations of the platform.…”

Section: Existing Workmentioning

confidence: 99%

“…The complex schemes in [17], [18] make them difficult for an on-line experimental validation. The studies [19], [21]- [23], [26], [27], [29], [31] need the information of obstacles. Only the simulations are presented in the studies [20], [24], [30].…”

Section: Existing Workmentioning

confidence: 99%

“…Properties of the EKF-FTIC are addressed by the following theorem. Theorem 1: Applying the EKF-FTIC (26), (27), (29)-(32) to the system (1)-(2) with the uncertainties (33) satisfying the inequality (34) and the output scaling factors satisfying the inequality (35), the operating point exponentially converges to the sliding surface (4), satisfying the convex set D :…”

Section: Fuzzy Tracking Incremental Controlmentioning

confidence: 99%

See 1 more Smart Citation

On-Line Obstacle Detection, Avoidance, and Mapping of an Outdoor Quadrotor Using EKF-Based Fuzzy Tracking Incremental Control

Hwang

Lai

2019

IEEE Access

View full text Add to dashboard Cite

The unknown two-dimensional obstacles, including different shapes, numbers, locations, and in the vicinity of the planning path for an outdoor quadrotor, are on-line detected by a 360-degree Li-Dar Sweep Scanse. Based on the comparison among detected triangles containing extreme points of each obstacle, the central orientation of the triangle with the maximum detected distance and the angle larger than the specific threshold is assigned to fly a specific distance for obstacle avoidance. Nevertheless, the quadrotor probably encounters a dead end and is then navigated to a specific altitude for the continuous procedure of obstacle detection. By on-line detection at different waypoints for the same obstacle, an obstacle mapping is also established by all the obstacle detections. Since the sensing signals of an outdoor quadrotor are easily affected by stochastic noise and the quadrotor is operated in ''x'' configuration, an extended Kalman filter based fuzzy tracking incremental control (EKF-FTIC) is employed to simultaneously execute the path tracking, obstacle avoidance, and target approach. Finally, the experiments with 5 unknown obstacles in the neighborhood of a planning path connecting start point with end point validate the effectiveness and feasibility of the proposed method. INDEX TERMS Extended Kalman filter, fuzzy tracking incremental control, on-line obstacle detection, avoidance and mapping, quadrotor.

show abstract

“…Due to the underactuated characteristics and limited maneuverability of the quadrotor (Mahony et al, 2012), the trajectory with excessively large desired acceleration or velocity may not be accurately executed (Morrell et al, 2018). The positioning and mapping module also may fail during the overaggressive motion.…”

Section: Introductionmentioning

confidence: 99%

A real‐time quadrotor trajectory planning framework based on B‐spline and nonuniform kinodynamic search

Tang

Wang

Liu

et al. 2020

Journal of Field Robotics

View full text Add to dashboard Cite

Autonomous navigation of quadrotor is required by many application scenarios, such as exploration, search, and rescue. The trajectory planning algorithm is the core of autonomous navigation, which can undoubtedly greatly enhance the safety of flight.In this paper, a trajectory planning framework based on B-spline and kinodynamic search is proposed. This framework can be used for a limited-sensing quadrotor to plan safe and dynamically feasible trajectories in unknown environments, and the flight is safe and effective along with these trajectories. First, a B-spline based nonuniform kinodynamic (BNUK) search algorithm is proposed to generate dynamically feasible trajectories efficiently. The characteristics of nonuniform search make the generated trajectories safe and reasonable time-allocation. Then, a trajectory optimization method based on control point optimization is proposed. The trajectory generated by BNUK is further optimized by solving a quadratically constrained quadratic programming problem. The smoothness of the trajectory is improved, and the control cost is reduced. Extensive analysis and comparative

show abstract

Comparison of Trajectory Optimization Algorithms for High-Speed Quadrotor Flight Near Obstacles

Cited by 21 publications

References 10 publications

An open‐source system for vision‐based micro‐aerial vehicle mapping, planning, and flight in cluttered environments

An open‐source system for vision‐based micro‐aerial vehicle mapping, planning, and flight in cluttered environments

On-Line Obstacle Detection, Avoidance, and Mapping of an Outdoor Quadrotor Using EKF-Based Fuzzy Tracking Incremental Control

A real‐time quadrotor trajectory planning framework based on B‐spline and nonuniform kinodynamic search

Contact Info

Product

Resources

About