Kinect based intelligent wheelchair navigation with potential fields

Ozcelikors, Mustafa; Coskun, Aylin; Say, Girayhan; Yayan, Uğur; Akcakoca, Mehmet; Kilic, Islam

doi:10.1109/siu.2014.6830249

Cited by 5 publications

(3 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A gradient then forms between the target and the current position of the robot; this forms a gradient descent trajectory from the current position to the target, making potential field methods highly suited to real-time mobile robotics applications [30]. This method has been adopted for collision avoidance with smart powered wheelchairs [31,32]; road vehicle driving assistance based upon using potential fields to keep vehicles following lanes and utilizing a steering damping method has also been suggested [33].…”

Section: Methodology For the Dynamic Localized Adjustable Force Fieldmentioning

confidence: 99%

A Dynamic Localized Adjustable Force Field Method for Real-Time Assistive Non-Holonomic Mobile Robotics

Gillham

Howells

2015

International Journal of Advanced Robotic Systems

View full text Add to dashboard Cite

Providing an assistive navigation system that augments rather than usurps user control of a powered wheelchair represents a significant technical challenge. This paper evaluates an assistive collision avoidance method for a powered wheelchair that allows the user to navigate safely whilst maintaining their overall governance of the platform motion. The paper shows that by shaping, switching and adjusting localized potential fields we are able to negotiate different obstacles by generating a more intuitively natural trajectory, one that does not deviate significantly from the operator in the loop desired-trajectory. It can also be seen that this method does not suffer from the local minima problem, or narrow corridor and proximity oscillation, which are common problems that occur when using potential fields. Furthermore this localized method enables the robotic platform to pass very close to obstacles, such as when negotiating a narrow passage or doorway.

show abstract

Section: Methodology For the Dynamic Localized Adjustable Force Fieldmentioning

confidence: 99%

A Dynamic Localized Adjustable Force Field Method for Real-Time Assistive Non-Holonomic Mobile Robotics

Gillham

Howells

2015

International Journal of Advanced Robotic Systems

View full text Add to dashboard Cite

show abstract

“…Among them we can mention LiDAR, single fixed beam, multiple fixed beams or rotating lasers in order to detect obstacles and determine the profile of the environment of the electric wheelchair, notably in the work of Grewal [ 12 ], Arnay [ 13 ] or the SmartWheelchair project [ 14 ] and the iChair project [ 15 , 16 ]. Özcelikörs [ 17 ] and Mittal [ 18 ] used the Kinect camera system in order to obtain a 3D map of the frontal obstacles, as well as their size and position. Other approaches use video processing, whether in mono- or stereovision, in order to detect particular objects or dangerous situations [ 19 , 20 ].…”

Section: Introductionmentioning

confidence: 99%

Development of a New Negative Obstacle Sensor for Augmented Electric Wheelchair

Favey

Farcy

Donnez

et al. 2021

Sensors

View full text Add to dashboard Cite

Due to pathologies or age-related problems, in some disabled people, motor impairment is associated with cognitive and/or visual impairments. This combination of limitations unfortunately leads to an inability to move around independently. Indeed, their situation does not allow them to use a conventional electric wheelchair, for safety reasons, and for the moment there is no other technological solution providing safe movement capacity. This lack of access to an autonomous travel solution has the consequence of weakening the intellectual, personal, social, cultural and moral development, as well as the life expectancy, of the people concerned. In this context, our team is working on the development of an optoelectronic system that secures the displacement of electric wheelchairs. This is a large project that requires the development of several functionalities such as: the anti-collision of the wheelchair with its environment, the prevention of falls from the wheelchair on uneven levels, and the adaptation of the system mechanically and electronically to the majority of commercially available electric wheelchair models, among others. In this article, we introduce our solution for detecting dangerous height differences, also called “negative obstacles”, through the creation of a dedicated sensor. This sensor works by optical triangulation and can embed several laser beams in order to extend its detection zone. It has the particularity of being robust in direct sunlight and rain and has a sufficiently high measurement rate to be suitable for the displacement of electric wheelchairs. We develop an adapted algorithm, and point out compromises, in particular between the orientation of the laser beams and the maximal speed of the wheelchair.

show abstract

“…The PFA was first suggested in [22], and developed later by Khatib who used the PFA as a real-time obstacle avoidance approach for mobile robots [23]. Recently, it has been also used for collision avoidance with the IW [24,25]. The concept combines positive forces acting at the goal with negative forces emerging from obstacles; it then builds a gradient descent trajectory from the current position to a specific target, which is well suited to real-time mobile robot navigation.…”

Section: Introductionmentioning

confidence: 99%

Wheelchair Navigation System for Disabled and Elderly People

Kim

2016

Sensors

View full text Add to dashboard Cite

An intelligent wheelchair (IW) system is developed in order to support safe mobility for disabled or elderly people with various impairments. The proposed IW offers two main functions: obstacle detection and avoidance, and situation recognition. First, through a combination of a vision sensor and eight ultrasonic ones, it detects diverse obstacles and produces occupancy grid maps (OGMs) that describe environmental information, including the positions and sizes of obstacles, which is then given to the learning-based algorithm. By learning the common patterns among OGMs assigned to the same directions, the IW can automatically find paths to prevent collisions with obstacles. Second, it distinguishes a situation whereby the user is standing on a sidewalk, traffic intersection, or roadway through analyzing the texture and shape of the images, which aids in preventing any accidents that would result in fatal injuries to the user, such as collisions with vehicles. From the experiments that were performed in various environments, we can prove the following: (1) the proposed system can recognize different types of outdoor places with 98.3% accuracy; and (2) it can produce paths that avoid obstacles with 92.0% accuracy.

show abstract

Kinect based intelligent wheelchair navigation with potential fields

Cited by 5 publications

References 10 publications

A Dynamic Localized Adjustable Force Field Method for Real-Time Assistive Non-Holonomic Mobile Robotics

A Dynamic Localized Adjustable Force Field Method for Real-Time Assistive Non-Holonomic Mobile Robotics

Development of a New Negative Obstacle Sensor for Augmented Electric Wheelchair

Wheelchair Navigation System for Disabled and Elderly People

Contact Info

Product

Resources

About