Development of a New Negative Obstacle Sensor for Augmented Electric Wheelchair

Driving Assistance System with Obstacle Avoidance for Electric Wheelchairs

A system has been developed to convert manual wheelchairs into electric wheelchairs, providing assistance to users through the implemented algorithm, which ensures safe driving and obstacle avoidance. While manual wheelchairs are typically controlled indoors based on user preferences, they do not guarantee safe driving in areas outside the user's field of vision. The proposed model utilizes the dynamic window approach specifically designed for wheelchair use, allowing for obstacle avoidance. This method evaluates potential movements within a defined velocity space to calculate the optimal path, providing seamless and safe driving assistance in real time. This innovative approach enhances user assistance and safety by integrating state-of-the-art algorithms developed using the dynamic window approach alongside advanced sensor technology. With the assistance of LiDAR sensors, the system perceives the wheelchair's surroundings, generating real-time speed values within the algorithm framework to ensure secure driving. The model's ability to adapt to indoor environments and its robust performance in real-world scenarios underscore its potential for widespread application. This study has undergone various tests, conclusively proving that the system aids users in avoidance obstacles and ensures safe driving. These tests demonstrate significant improvements in maneuverability and user safety, highlighting a noteworthy advancement in assistive technology for individuals with limited mobility.

SWADAPT2: benefits of a collision avoidance assistance for powered wheelchair users in driving difficulty

PURPOSE

In France, tens of thousands of people use a wheelchair. Driving powered wheelchairs (PWCs) present risks for users and their families. The risk of collision in PWC driver increases with severity of disability and may reduce their independence to drive. The European ADAPT project has developed a robotic assistance add-on for PWCs to prevent collisions and improve their driving performance.

MATERIALS AND METHODS

The aim of the SWADAPT2 study is to assess the benefit of this robotic assistance add-on arranged on a Standard Quickie Salsa M2 PWC in a population of PWC drivers with neurological disorders and driving difficulties. Eighteen (18) participants tested the system on three circuits of increasing difficulty, with and without the robotic assistance add-on.

RESULTS

The benefit of the robotic assistance add-on was important especially on the more difficult circuits without impacting cognitive load or driving speed. The number of collisions was significantly reduced when using robotic assistance add-on from 2.16 to 0.36 on circuit 2 (p = 0.009) and from 7.3 to 1.33 on circuit 3 (p = 0.0009). Task load demand was not increased with the assistance.

CONCLUSION

Finally, this system seems to be indicated to assist and improve driving safety for PWC drivers in driving difficulty. Evaluation was performed in controlled environments; therefore, further evaluation in real-world scenarios is needed to reach technology readiness.

Classification of the Sidewalk Condition Using Self-Supervised Transfer Learning for Wheelchair Safety Driving

The demand for wheelchairs has increased recently as the population of the elderly and patients with disorders increases. However, society still pays less attention to infrastructure that can threaten the wheelchair user, such as sidewalks with cracks/potholes. Although various studies have been proposed to recognize such challenges, they mainly depend on RGB images or IMU sensors, which are sensitive to outdoor conditions such as low illumination, bad weather, and unavoidable vibrations, resulting in unsatisfactory and unstable performance. In this paper, we introduce a novel system based on various convolutional neural networks (CNNs) to automatically classify the condition of sidewalks using images captured with depth and infrared modalities. Moreover, we compare the performance of training CNNs from scratch and the transfer learning approach, where the weights learned from the natural image domain (e.g., ImageNet) are fine-tuned to the depth and infrared image domain. In particular, we propose applying the ResNet-152 model pre-trained with self-supervised learning during transfer learning to leverage better image representations. Performance evaluation on the classification of the sidewalk condition was conducted with 100% and 10% of training data. The experimental results validate the effectiveness and feasibility of the proposed approach and bring future research directions.

An Analytical Modelling of Demand for Driving Torque of a Wheelchair with Electromechanical Drive

This study aimed at analysing the influence of the position of the centre of gravity variability and the movement velocity on the demand for a torque and work time coverage of a wheelchair with an electromechanical drive. The variable parameter in the study was the configuration of the wheelchair, namely changes in the position of the batteries which changed the weight distribution. An analytical model describing the demand for torque was used in the analysis. The set of equations was introduced into the numerical calculation software. Simulations were carried out which allowed it to analyse selected parameters of the wheelchair dynamics. An increase in the torque demand was observed due to the increase in the mass of the system from 427.7 N to 533.1 N, ranging from 6.1% to 31.6% at the simulated velocity v7 = 4.2 m/s. The increase in the demand for torque due to the increase in velocity of the wheelchair from v2 = 1.05 m/s to v7 = 4.2 m/s ranged from 25.9% to 31.6% compared to the reference velocity v1 = 0.525 m/s. The centre of gravity of the wheelchair structure localization has a non-linear impact on the analysed values. At the same time, it was not possible to define its nature—this issue remains open and requires further research.