Reactive avoidance using embedded stereo vision for MAV flight

Oleynikova, Helen; Honegger, Dominik; Pollefeys, Marc

doi:10.1109/icra.2015.7138979

Cited by 91 publications

(67 citation statements)

References 19 publications

(23 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If obstacle avoidance is the main focus for the motion planner, less expensive representation and planning algorithms are possible, even in highly cluttered scenes. Representations that are polar in nature, which match the polar angular resolution of the depth sensors 14,15 can be used to reflect the emphasis on objects in the proximity of the vehicle. There has been work 16 that used the advantage of such a polar perspective implementation by fusing depth data from two on-board stereo pairs in a cylindrical inverse range map centered on the vehicle.…”

Section: Related Workmentioning

confidence: 99%

Stereo vision-based obstacle avoidance for micro air vehicles using an egocylindrical image space representation

Fragoso

Matthies

2016

SPIE Proceedings

View full text Add to dashboard Cite

Micro air vehicles which operate autonomously at low altitude in cluttered environments require a method for onboard obstacle avoidance for safe operation. Previous methods deploy either purely reactive approaches, mapping low-level visual features directly to actuator inputs to maneuver the vehicle around the obstacle, or deliberative methods that use on-board 3-D sensors to create a 3-D, voxel-based world model, which is then used to generate collision free 3-D trajectories. In this paper, we use forward-looking stereo vision with a large horizontal and vertical field of view and project range from stereo into a novel robot-centered, cylindrical, inverse range map we call an egocylinder. With this implementation we reduce the complexity of our world representation from a 3D map to a 2.5D image-space representation, which supports very efficient motion planning and collisionchecking, and allows to implement configuration space expansion as an image processing function directly on the egocylinder. Deploying a fast reactive motion planner directly on the configuration space expanded egocylinder image, we demonstrate the effectiveness of this new approach experimentally in an indoor environment.

show abstract

Section: Related Workmentioning

confidence: 99%

Stereo vision-based obstacle avoidance for micro air vehicles using an egocylindrical image space representation

Fragoso

Matthies

2016

SPIE Proceedings

View full text Add to dashboard Cite

show abstract

“…The reported power consumption of the module is 5 Watts. In [16], the same authors extend their module by implementing a reactive based collision avoidance method for MAVs and tested it in an outdoor environment.…”

Section: Related Workmentioning

confidence: 99%

On Fpga Based Acceleration of Image Processing in Mobile Robotics

Petr

Faigl

2015

APP

View full text Add to dashboard Cite

Abstract. In visual navigation tasks, a lack of the computational resources is one of the main limitations of micro robotic platforms to be deployed in autonomous missions. It is because the most of nowadays techniques of visual navigation relies on a detection of salient points that is computationally very demanding. In this paper, an FPGA assisted acceleration of image processing is considered to overcome limitations of computational resources available on-board and to enable high processing speeds while it may lower the power consumption of the system. The paper reports on performance evaluation of the CPU-based and FPGA-based implementations of a visual teach-and-repeat navigation system based on detection and tracking of the FAST image salient points. The results indicate that even a computationally efficient FAST algorithm can benefit from a parallel (low-cost) FPGA-based implementation that has a competitive processing time but more importantly it is a more power efficient.

show abstract

“…Recent work has shown that obstacle detection for reactive collision avoidance is possible at high speeds on a CPU (68) running onboard a fixed-wing aircraft traveling at up to 14 m/s. However, maintaining a complete map for longer-range planning is computationally expensive, and as a result, most mapping algorithms have been executed using either offboard resources (50) or onboard specialized hardware, such as GPUs (53) or field-programmable gate arrays (69). These mapping algorithms can be combined with trajectory-generation methods to enable autonomous navigation through cluttered environments (49,53,68).…”

Section: State Estimation and Perceptionmentioning

confidence: 99%

Autonomous Flight

Tang

Kumar

2018

Annu. Rev. Control Robot. Auton. Syst.

View full text Add to dashboard Cite

This review surveys the current state of the art in the development of unmanned aerial vehicles, focusing on algorithms for quadrotors. Tremendous progress has been made across both industry and academia, and full vehicle autonomy is now well within reach. We begin by presenting recent successes in control, estimation, and trajectory planning that have enabled agile, highspeed flight using low-cost onboard sensors. We then examine new research trends in learning and multirobot systems and conclude with a discussion of open challenges and directions for future research. 6.1

show abstract

Reactive avoidance using embedded stereo vision for MAV flight

Cited by 91 publications

References 19 publications

Stereo vision-based obstacle avoidance for micro air vehicles using an egocylindrical image space representation

Stereo vision-based obstacle avoidance for micro air vehicles using an egocylindrical image space representation

On Fpga Based Acceleration of Image Processing in Mobile Robotics

Autonomous Flight

Contact Info

Product

Resources

About