The safety of intelligent wheelchairs equipped with navigation functionality is critical, especially in complex indoor environments. Traditional 2D LiDAR sensors are limited in their ability to detect three-dimensional obstacles, which can result in potential hazards for users. While RGB-D cameras are capable of reconstructing 3D scenes, their performance is significantly hindered by uneven lighting conditions that are common in indoor spaces. Depth-dependent light sources, which these cameras rely on, cannot consistently provide accurate scene reconstruction in such environments, thereby posing safety risks for elderly or disabled users. This paper proposes a novel intelligent nursing wheelchair that integrates 3D LiDAR technology to address these challenges. Unlike 2D LiDAR sensors or RGB-D cameras, the 3D LiDAR sensor can effectively detect irregular and height-differentiated obstacles in complex indoor environments without relying on uniform lighting. Experimental mapping of indoor environments demonstrates that 3D LiDAR can accurately identify and navigate around three-dimensional obstacles, significantly improving the wheelchair’s ability to operate safely in dynamic indoor spaces. This development enhances the overall safety and mobility of elderly users, reducing the risk of accidents and improving their independence.

Addressing the challenge that existing assisted bathing robots face in performing compliant scrubbing on the complex surface contours of the human body, this study proposes a compliance control strategy designed to enhance scrubbing effectiveness and precision. The strategy is implemented using a six-degree-of-freedom robotic arm, which integrates machine vision and force-tactile perception to adapt to varying body contours dynamically. First, the stress exerted on the skin during scrubbing is analyzed to ensure user safety and comfort. Traditional impedance control is then optimized using a force-error differential method weighted by the Logistic growth curve, and its performance is evaluated through simulations in Simulink. To further improve force-tracking accuracy, an enhanced impedance controller based on variable-universe fuzzy control is developed and compared against a conventional non-fuzzy-control approach. Finally, human experiments are conducted to validate real-world performance. The trajectory error remains within ±3mm, while the contact force error is constrained to ±0.5N. Experimental results confirm that the proposed strategy achieves effective scrubbing performance with high precision, demonstrating significant improvements over traditional methods.

This paper presents a structured methodology for the rapid development of modular robots designed to capture and address user requirements in care settings. While co-design is widely applied in robotics, existing approaches often lack detailed, step-by-step frameworks. This study employs a Design Research (DR) methodology to establish a comprehensive process, demonstrated through the development of a modular robot, X, in collaboration with care home users during the COVID-19 pandemic.

The proposed methodology follows a seven-step structured process that systematically captures and clusters user requirements while integrating iterative feedback to refine modular robot variations. Design for Additive Manufacturing (DfAM) is used to expedite prototype production, enabling rapid iterations informed by user input. This approach enhances the adaptability of care and service robots, ensuring they align more effectively with user needs by fostering active engagement throughout the design process.

This research outlines the complete methodology, the tools developed to capture and assess user requirements, and the evaluation of progress and insights gained. The findings provide a replicable framework for participatory robot design in care environments, contributing to the broader research community.

The sustainability of healthcare systems is increasingly challenged by aging populations, chronic diseases, and limited resources. Mobile robotics offers a potential solution by enhancing efficiency, supporting healthcare professionals, and improving patient care. This study analyzes survey responses from 47 healthcare professionals across 17 occupations in Finland to identify applications for mobile robotics in healthcare environments. The results highlight the following key areas where mobile robotics could be utilized: transport and logistics, patient and caregiver assistance, remote monitoring and other tasks. Logistical applications, such as transporting medicines, medical supplies, and food trolleys, were identified as the most crucial, as automation could reduce manual labor and free up time for direct patient care. Respondents also saw potential in patient support, remote monitoring, and administrative tasks. While mobile robotics is still underutilized in healthcare, technological advancements and decreasing costs could expand their applications. As mobile robots become more common, their acceptance among professionals and patients will grow, enabling them to serve as multifunctional assistive tools in healthcare.

The development and application of feeding-assistance robots represent an effective approach to help elderly individuals with disabilities achieve autonomous eating. This paper introduces a feeding-assistance robot specifically designed for elderly individuals with disabilities in home care environments who are unable to eat independently due to limb dysfunction. We present a vision-based interactive feeding-assistance robot featuring six degrees of freedom (6-DOF), incorporating a flexible design in partial joints to form a rigid-flexible hybrid robotic arm. Additionally, it integrates a vision interaction system to achieve reliable feeding assistance for elderly individuals with disabilities in constrained home spaces. Practical applications demonstrate that the vision-interaction-based meal-assistance robot is more intelligent compared to traditional button-interaction-based systems. Notably, our independently improved visual interaction algorithm achieves a 25% increase in recognition speed per second and a 30% reduction in algorithm size while maintaining high accuracy, compared to the original algorithm. Experimental results confirm that the self-designed meal-assistance robot effectively supports elderly individuals with disabilities in daily eating activities.