Robotics: New skin-like sensors fit almost everywhere

Researchers from the Munich Institute of Robotics and Machine Intelligence (MIRMI) at the Technical University of Munich (TUM) have made a significant breakthrough in the field of soft sensors. They have developed an automated process for creating universal measurement cells that can be attached to nearly any object. This innovation holds great potential for applications in robotics and prosthetics.

According to Sonja Groß, a researcher from MIRMI, understanding and sensing our environment is crucial for effective interaction with it. The shape of objects plays a vital role in determining how we can perform tasks, while their physical properties, such as hardness and flexibility, influence our ability to grasp and manipulate them.

In the fields of robotics and prosthetics, achieving sensorimotor skills similar to those of a human hand is a major goal. Force and torque sensors are already integrated into many robotic systems, providing valuable feedback on interactions with the surroundings. However, traditional sensors have limitations in terms of customization and attaching them to arbitrary objects. Until now, there was no process for producing sensors suitable for objects of various shapes and sizes.

The research team, led by Sonja Groß and Diego Hidalgo, has developed a new framework for soft sensors, which they recently presented at the ICRA robotics conference in London. Their approach involves using a soft, skin-like material that wraps around objects. They have also created a software-driven production process that automates the manufacturing of these sensors. A 3D printer is employed to inject a conductive black paste into liquid silicone, which then hardens. The resulting sensors exhibit changes in electrical resistance when squeezed or stretched, providing information about compression and stretching forces. This data helps understand interactions with objects and enables precise control of artificial hands interacting with the environment.

What sets this research apart is that the sensors embedded in the silicone adjust to the surface they are applied to, such as fingers or hands, while still delivering precise and valuable data for interaction with the environment.

Prof. Sami Haddadin, the Executive Director of MIRMI, believes that integrating these soft, skin-like sensors into 3D objects opens up new possibilities for advanced haptic sensing in artificial intelligence. The real-time feedback from these sensors on compressive forces and deformations expands the perception range of objects or robotic hands, enabling more sophisticated and sensitive interactions. Haddadin envisions this breakthrough revolutionizing industries like robotics, prosthetics, and human-machine interaction, making it possible to create wireless and customizable sensor technology for a wide range of objects and machines.

Posted in

Aihub Team

Leave a Comment





SK Telecom outlines its plans with AI partners

SK Telecom outlines its plans with AI partners

Razer and ClearBot are using AI and robotics to clean the oceans

Razer and ClearBot are using AI and robotics to clean the oceans

NHS receives AI fund to improve healthcare efficiency

NHS receives AI fund to improve healthcare efficiency

National Robotarium pioneers AI and telepresence robotic tech for remote health consultations

National Robotarium pioneers AI and telepresence robotic tech for remote health consultations

IBM’s AI-powered Mayflower ship crosses the Atlantic

IBM’s AI-powered Mayflower ship crosses the Atlantic

Humans are still beating AIs at drone racing

Humans are still beating AIs at drone racing

How artificial intelligence is dividing the world of work

How artificial intelligence is dividing the world of work

Global push to regulate artificial intelligence

Global push to regulate artificial intelligence

Georgia State researchers design artificial vision device for microrobots

Georgia State researchers design artificial vision device for microrobots

European Parliament adopts AI Act position

European Parliament adopts AI Act position

Chinese AI chipmaker Horizon endeavours to raise $700M to rival NVIDIA

Chinese AI chipmaker Horizon endeavours to raise $700M to rival NVIDIA

AI Day: Elon Musk unveils ‘friendly’ humanoid robot Tesla Bot

AI Day: Elon Musk unveils ‘friendly’ humanoid robot Tesla Bot

AI and Human-Computer Interaction: AI technologies for improving user interfaces, natural language interfaces, and gesture recognition.

AI and Data Privacy: Balancing AI advancements with privacy concerns and techniques for privacy-preserving AI.

AI and Virtual Assistants: AI-driven virtual assistants, chatbots, and voice assistants for personalized user interactions.

AI and Business Process Automation: AI-powered automation of repetitive tasks and decision-making in business processes.

AI and Social Media: AI algorithms for content recommendation, sentiment analysis, and social network analysis.

AI for Environmental Monitoring: AI applications in monitoring and protecting the environment, including wildlife tracking and climate modeling.

AI in Cybersecurity: AI systems for threat detection, anomaly detection, and intelligent security analysis.

AI in Gaming: The use of AI techniques in game development, character behavior, and procedural content generation.

AI in Autonomous Vehicles: AI technologies powering self-driving cars and intelligent transportation systems.

AI Ethics: Ethical considerations and guidelines for the responsible development and use of AI systems.

AI in Education: AI-based systems for personalized learning, adaptive assessments, and intelligent tutoring.

AI in Finance: The use of AI algorithms for fraud detection, risk assessment, trading, and portfolio management in the financial sector.

AI in Healthcare: Applications of AI in medical diagnosis, drug discovery, patient monitoring, and personalized medicine.

Robotics: The integration of AI and robotics, enabling machines to perform physical tasks autonomously.

Explainable AI: Techniques and methods for making AI systems more transparent and interpretable

Reinforcement Learning: AI agents that learn through trial and error by interacting with an environment

Computer Vision: AI systems capable of interpreting and understanding visual data.

Natural Language Processing: AI techniques for understanding and processing human language.