Sestosenso develops technologies for next generations of collaborative robots capable of self-adapting to different time-varying operational conditions and capable of safe and smooth adaptation from autonomous to interactive when human intervention is required either for collaboration or training/teaching. The project proposes a new sensing technology from the hardware and up to the cognitive perception and control levels, based on networks of embedded proximity and tactile sensors on the robot body, providing a unified proxy-tactile perception of the environment, required to control the robot actions and interactions, safely and autonomously. Within the project, the same technologies are also applied to wearable devices (like exoskeletons) to provide the user with better spatial awareness and to enforce safety in critical human-robot interactive tasks.
Sestosenso is motivated by the growing industrial need to integrate production line operators with robots, equipment, and site-specific factory information according to the principles of Industry 4.0 (interconnected automation) and the forthcoming Industry 5.0 (humanization and re-use of resources). This leads to the need for robot systems that could operate and interact safely with a limited need of out-of-the-robot infrastructure, and possibly reducing robot setup times and costs thus increasing the flexibility of the shopfloor configuration. Also, Sestosenso is motivated by the need of adopting robots for medium-small production/service facilities, as well as for domestic applications, with limited possibility of infrastructure investments, and finally by the need of developing robot applications in poorly structured and rapidly changing environments or outdoor. To this aim, the goal of Sestosenso is to develop self- standing robot systems with high cognitive skills with the capability to execute operations involving objects and/or humans in unstructured and dynamic environments without the need of specific external infrastructures.
Sestosenso is grounded on the idea of integrating, over the robot1 body, networks of discrete miniaturized proximity sensors together with tactile sensors. This solution leads to ProxySKIN, a skin-like sensory system providing a complete and seamless proxy-tactile perception of the robot surroundings where proximity and tactile information can be intended as a new extended sensing mode (skin-vision2).
This idea opens new challenges for the development of robots with high cognitive capabilities and motivates:
- the study of perception and control methods for autonomous and human driven robot operations
- the implementation of new types of robot tasks exploiting the joint proxy-tactile feedback
- the development of techniques to enforce the safety in human-robot interaction (HRI)
- the development of techniques for self-localization of the robot and models for robot-centric environment representation
- the development of innovative manufacturing solutions for the construction of large area distributed sensors enabling the large scale implementation of ProxySKIN sensors.
The goal of Sestosenso is to develop and demonstrate (by means of 3 industrially relevant Use Cases) the capabilities of self-standing robot systems using proxy-tactile feedback provided by ProxySKIN sensors3. In a top-down view, the robots demonstrate the possibility:
- to work safely and in close cooperation with humans to perform assembly tasks in a dynamic and narrow environment (Use Case 1)
- to work outdoor in presence of unmodelled and delicate obstacles (plants) in cooperation with a human wearing a sensorized exoskeleton to enhance human capabilities and HRI safety (Use Case 3)
- to exploit human gesture/touch guided teaching for learning the handling of large heavy and bulky objects, in single or dual-arm configurations, using whole-arm-manipulation (WAM) modes thus fully exploiting the robot lifting capabilities (Use Case 2)
Then, robots are expected in general:
- to adapt their behaviour in presence of changes in the nearby space
- to properly react to and control the contact forces acting all over the robot body
- to recognize human gestures and touch to engage safe human robot interaction.
- to implement cost effective and versatile design methods and fabrication processes for conformable and possibly stretchable multimodal proxy-tactile large area sensors
- to develop sensors which could be integrated on different types of robots and with different shapes
- to explore new large area sensor concepts in order to target cheaper and more reliable components
Sestosenso addresses the development of technologies devised for industrial and service applications and targets the development of human-centred technologies. The outcomes of the project follow the expected pathway on innovation in AI, data and robotics, and targets the development of robot technologies appealing for real industrial applications. The project aims at serving the needs of several types of European industries, including manufacturing, logistics, agriculture, but potentially addressing many others. Sestosenso creates a specific robot technology portable over different application domains and multiple robot and exoskeletons platforms, capable of high cognitive skills to engage safe interaction with common users. This approach enables a wider introduction of collaborative robots in large, medium and small production companies with the potential of creating new production systems bridging to employment, inclusion and workers wellbeing.
During the development of hardware and software solutions, it is expected that technologic achievements attract the attention of scientists, technologists, and practitioners and foster the formation of young talents. Specific scientific impact is expected in the domains of advanced skin vision, hybrid printed and standard electronics on potentially curved surfaces, wireless real-time networks and distributed optical tactile transducers