• autonomous navigation
• human-robot interaction
• intuitive programming

Today, robots frequently operate in crowded, dynamic environments, often near humans. Within industrial settings, there is a growing demand for robots to collaborate and coexist with human workers, driven by the evolution from Industry 4.0 to Industry 5.0. Various approaches have been explored to achieve this synergy between robots and humans. In industrial scenarios, the robots encountered can range from traditional robotic arms on assembly lines to autonomous mobile robots navigating warehouse floors. These robots play a crucial role in enhancing efficiency and productivity across different tasks within the industrial landscape, making them integral elements of the Industry 4.0 and 5.0 transformations that are reshaping the way we manufacture and operate.

My doctoral research explores these phenomena, starting with a specific case study: a locomanipulation task for logistics in industrial plants. This task involves the coordination of robots for the movement and manipulation of objects within the facility. The challenges faced in this scenario include not only planning and manipulation but also encompass aspects such as real-time decision-making, adaptability to dynamic environments, and human-robot interaction.

Traditional robotic techniques have been applied to address the planning and manipulation challenges. However, as the industrial landscape evolves, new challenges emerge, necessitating innovative solutions.

One such challenge is the need for robots to navigate efficiently through crowded spaces while ensuring the safety of both the robots themselves and human workers. As a response, my research focused on developing an efficient 2D LIDAR-based map updating mechanism to enhance localization and navigation in dynamic environments. Additionally, I proposed two distinct planners to address safety concerns in human-centric environments: a risk-aware planning approach utilizing collision probability maps and a context-aware planner leveraging 3D human features to plan a human-aware trajectory. These planners aimed to promote a harmonious coexistence between robots and human workers within industrial settings, emphasizing safety and adaptability in their interactions.

Recognizing that the transition towards Industry 5.0 is not just about maximizing productivity but also about integrating human values and ingenuity with technological advancements, a wide range of methods and techniques are currently in development to endow robots with cognitive abilities akin to humans, enabling them to achieve a high level of autonomy.

The core of this progress lies in developing robotic partners that can mirror human reasoning and skills, fostering a more natural and effective collaboration with humans. This advancement enhances robots' capability to autonomously make decisions and solve complex problems like their human counterparts.

In this context, a growing emphasis is on developing intuitive interfaces for programming robotic systems. The goal is to create robots that think and act in ways paralleling human expertise, facilitating a straightforward and efficient transfer of skills. These advancements are required to make robotic technology accessible to a broader range of users, regardless of their expertise in robotics.

In response to this imperative, my research includes innovative approaches, exemplified by the design, development, and testing of a voice visual user interface for robot programming without a coding system. This initiative empowers users by simplifying the programming process and enabling intuitive robot control through voice and visual interactions, reducing the entry barrier for non-expert users and fostering a more accessible and user-friendly interface.

Finally, always remaining in the context of developing intuitive, accessible robotics and adaptive systems, another significant aspect of my research involves investigating saliency for learning sensory-motor contingencies in loco-manipulation tasks. Thus, I proposed a framework for a robot to learn multiple Sensory-Motor Contingencies from human demonstrations and reproduce them. Sensory-motor contingencies are a concept that describes the intelligent behavior of animals and humans about their environment. I aimed to transfer this behavior to the robot through human demonstrations, enhancing its ability to learn and adapt to its environment during complex loco-manipulation tasks, ultimately contributing to improved autonomy.

In conclusion, the journey from addressing challenges in dynamic environments to enhancing robotic autonomy is marked by a progression from classical techniques to contemporary approaches that make robotics accessible even to non-experts. This trajectory reflects the symbiotic relationship between robots and humans, where the fusion of their strengths results in more capable, adaptable systems programmed through intuitive systems and interfaces.