Technology Strategy

Technology Strategy Consulting

Key Technology Development Questions for Sensors in Soft Robotics

In the rapidly evolving field of soft robotics, the integration of advanced sensor technologies is essential to enhance flexibility, resilience, and control precision. This blog explores the critical technological challenges facing soft robotics today, addressing key questions related to sensor limitations, integration complexities, and the future of autonomous functionality. Drawing from recent advancements and bio-inspired designs, we highlight how emerging materials, AI-driven data interpretation, and multimodal sensing capabilities are shaping the next generation of soft robots. Through this discussion, we also point out the potential of soft robotics across applications in healthcare, manufacturing, and beyond, as these systems become increasingly adaptable and capable of human-like dexterity.

From a technology perspective, there are several key questions to address when considering sensing in soft robotics. These questions are designed to evaluate the current capabilities, challenges, and future opportunities in this emerging field. In this blog, we strive to answer 7 key questions based on information presented in the previous 3 blogs on soft robots, titled; Bio-Inspired Soft Robots: Transforming Medical Implants with E-Skin and Artificial Muscles, Soft Robotics: Advances, Challenges, and Future Applications in Manipulation, Exploration, and Healthcare, and Soft Robotics Sensing: Advancing Flexible Automation with Integrated Sensor Technologies. The sources used here are from the publications cited in the previous blogs and their sources.

Sensor Limitations Hinder Soft Robotics

Figure 1. Current sensor technologies in soft robotics are hindered by limited durabilityintegration complexitycontrol precision challenges, and sensor suitability issues, which restrict their effectiveness in various applications.

Several key limitations of current sensor technologies within the soft robotics field are noted:

  • Limited Durability: Soft robots are constructed from materials like silicone, hydrogels, and flexible polymers that allow for flexibility and safe interaction with delicate objects. [1] However, this comes at the expense of durability in demanding industrial environments where strength and resilience are critical. [2, 3] Soft materials are more susceptible to wear and tear under harsh conditions and heavy loads, potentially restricting their applications in certain fields. [2, 3]

  • Integration Complexity: Integrating sensors into soft robots while preserving flexibility and mechanical compliance is a significant challenge. [35] Advanced sensors and control systems required for precise movement and response in soft robots introduce additional complexity. [6] These systems require significant research and development, posing a barrier to widespread adoption. [6] Designing small, deformable, but high-performance parts, especially sensors and actuators, presents challenges for soft robot design. [5] These parts demand extensive research and skilled developers to produce, further complicating the integration process. [5]

  • Control Precision: The compliant nature of soft robots makes achieving precise control and motion difficult. [4, 7] Soft materials deform and stretch, making it challenging to attain the same level of precision and repeatability as traditional rigid robots. [4] This lack of precision can be a major drawback in applications demanding high accuracy, such as manufacturing and surgical procedures. [4] Soft robots lack the discrete joints and rigid links found in traditional robots, requiring new control schemes specific to their high degrees of freedom and continuous deformation capabilities. [8, 9]

  • Sensor Suitability: Conventional sensors designed for rigid structures are not always easily integrated into the compliant bodies of soft robots. [10] Soft robots theoretically have infinite degrees of freedom, requiring sensors with sufficient motion range, accuracy, and resolution to measure continuous posture changes and reconstruct the overall structure. [10] Researchers must carefully evaluate sensor performance metrics, considering object detection, deformation type, nonlinearity, hysteresis, and cost when choosing the appropriate technology. [11]

While resistive, capacitive, optical, and magnetic sensors offer promising capabilities for soft robotics, each technology has its own drawbacks that need to be considered. [10, 11]

Resistive sensors are a low-cost solution for proprioceptive soft robots, but suffer from hysteresis, potentially affecting the accuracy of dynamic movement measurements. [11]

Capacitive sensors offer a large dynamic range, fast response, and excellent linear range and sensitivity. However, they can be susceptible to contaminants, proximity effects, and mechanical perturbation, leading to variations in quantitative pressure measurements. [12, 13]

Optical fiber sensors provide promising accuracy in measuring continuous deformation, but the readout terminal can be bulky, limiting mobile applications. [14] These systems, while commercially available, are expensive, hindering widespread adoption. [15]

Magnetic sensors are compact, low-cost, and not prone to hysteresis. However, they are vulnerable to interference from external metals or magnets and can experience crosstalk with nearby magnetic sensors unless carefully shielded. [15]

Overall, current sensor technologies do not always provide the precision, flexibility, and resilience required for all real-world soft robotic applications. Ongoing research and innovation in materials science, robotics engineering, and control systems are needed to overcome these limitations and unlock the full potential of soft robotics.

Integrating Sensors in Soft Robots

Figure 2. Integrating sensors into soft robots effectively requires the use of flexible materials and advanced manufacturing techniques to maintain their mechanical compliance while ensuring accurate data collection.

Q#2: How can sensors be better integrated into soft robots without compromising their flexibility and mechanical compliance?

The key issue is the challenge of integrating sensors into soft robots while preserving flexibility and accurate data collection [13]. Emerging integration methods prioritize using flexible and stretchable materials, and advanced manufacturing techniques like 3D printing are critical for seamless sensor integration.

  • Material Science Advancements: The use of elastomers is critical in soft robot construction due to their ability to withstand significant deformation without permanent damage [4]. These materials allow sensors to be integrated directly into the robot’s structure without hindering flexibility [4].

  • 3D Printing and Additive Manufacturing: These techniques enable the creation of intricate and customized soft robot designs [4, 5]. This customization extends to embedding sensors within the robot’s structure, providing real-time feedback for precise control [1, 5].

Beyond the physical integration of sensors, the sources highlight several approaches that ensure minimal interference with movement and accurate data collection:

  • Soft Sensors: The development of soft sensors specifically designed for flexible robots is an important area of research [3, 6]. These sensors are built using materials and structures that can bend and stretch with the robot’s body [6].

  • Distributed Proprioception: This approach involves embedding multiple sensors throughout the robot’s body [7]. By distributing sensors, researchers can gain a more complete understanding of the robot’s shape and movement, leading to more accurate control [79].

  • Bio-Inspired Designs: There is an emphasis around the importance of looking to nature for inspiration in soft robot design [5, 10]. For example, mimicking the structure and function of human skin, with its distributed network of sensors, can provide insights into developing more sensitive and robust soft robotic sensing systems [1113].

  • Advanced Control Systems: The key issue here is the development of sophisticated control systems is crucial for processing the data collected by sensors and translating it into precise movements [1, 14, 15]. These control systems must be able to handle the complexities of soft robot dynamics, including nonlinearity and underactuation [9, 15].

There are various specific sensor technologies and their integration challenges. Resistive sensors, for example, are cost-effective but prone to hysteresis, while capacitive sensors offer good sensitivity but can be affected by environmental factors [8, 16]. Optical sensors, such as fiber Bragg grating (FBG) sensors, are accurate but can be bulky [17, 18]. Magnetic sensors are compact and robust but susceptible to interference [17, 19].

Addressing these challenges requires ongoing research and development. Researchers are exploring new materials with enhanced properties, such as conductive liquids and inks that can be integrated into complex 3D printed structures [20, 21]. They are also developing advanced algorithms and machine learning techniques to improve data processing and control, enabling soft robots to adapt to changing environments and tasks [22, 23, 24, 25].

There is an emphasis that successful sensor integration in soft robotics is a multidisciplinary effort. It requires collaboration between material scientists, robotics engineers, and computer scientists to develop innovative solutions that push the boundaries of what soft robots can achieve.

Advancements Needed for Autonomous Soft Robots

Figure 3. To achieve true autonomy in soft robots, advancements are needed in miniaturized, low-power sensors, compact power sources, and efficient wireless technologies, alongside dedicated control systems and collaborative innovation among various scientific disciplines.

Q#3: What breakthroughs are needed in sensor miniaturization and power efficiency for autonomous soft robots?

The key is to highlight the need for smaller, more efficient sensors for soft robots to become truly autonomous. While the sources don’t explicitly discuss specific sensor miniaturization breakthroughs, they do address power consumption and the limitations of current power sources. Here’s a breakdown of the key areas where innovation is needed:

  • Energy Harvesting: There is no direct discussion on energy harvesting for soft robots in our targeted sources. However, integrating energy harvesting technologies like solar panels [1, 2], fuel cells [2, 3], or triboelectric nanogenerators [46] could potentially reduce reliance on bulky batteries and enable longer operation times.

  • Low-Power Sensors: The power requirements differ for different soft robot actuators and sensors [7]. The pneumatic actuators and ionic polymer–metal composites exhibit the lowest driving voltage, while dielectric elastomer actuators demand the highest. This highlights the need to develop sensors with equally low power requirements to match the capabilities of these efficient actuators. This could involve designing new sensing materials and architectures that minimize energy consumption.

  • Wireless Technologies: Discussions highlight the importance of wireless operation for soft robots, particularly in medical applications [8]. These include examples of wirelessly controlled soft robots using electromagnetic power transmission [9, 10] and inductive coils for transmitting sensing signals [9]. To enhance autonomy, further advancements in wireless power transfer and data communication are needed. This could involve developing more efficient and miniaturized wireless modules specifically designed for integration into soft robots.

  • Compact Power Sources: The bulky nature of current power sources, especially for fluidic soft robots, limits their mobility and autonomy [3, 11]. Research into miniaturized batteries and other compact energy storage solutions is crucial to enable untethered operation. This could involve exploring new battery chemistries and designs or developing alternative power sources like micro-fuel cells.

There is emphasis on the importance of dedicated control systems for soft robots [12]. These systems would need to be miniaturized and energy-efficient to support autonomous operation. Additionally, the development of stable interfaces between soft and rigid components, such as actuators, controllers, sensors, and power units, is crucial [13].

Overall, achieving true autonomy in soft robotics will require a multi-pronged approach. It will involve developing new materials, miniaturizing existing technologies, and designing systems optimized for low power consumption. The sources suggest that collaboration between material scientists, robotics engineers, and computer scientists is essential to overcome these challenges and bring about the next generation of autonomous soft robots.

Using AI and Machine Learning to Improve Soft Robot Control

Figure 4. AI and machine learning enhance soft robot control by enabling adaptive learning, improving precision, processing multimodal data, and optimizing closed-loop systems, leading to more versatile and intelligent robotic applications.

Q#4: What role can machine learning and AI play in improving sensor data interpretation and control in soft robotics?

There are various examples of how machine learning and AI can be applied to enhance sensor data interpretation and control within soft robotics.

AI and machine learning algorithms are particularly well-suited for addressing the challenges posed by the complex, nonlinear data generated by soft robot sensors. These technologies allow soft robots to:

  • Learn from Experience: Unlike traditional rigid robots, which rely on pre-programmed instructions, AI-powered soft robots can learn from their interactions with the environment and adapt their behaviors accordingly [13]. This makes them more effective in performing complex tasks and responding to unexpected situations.

  • Improve Precision: Soft robots struggle to achieve the same level of precision as their rigid counterparts due to their deformable nature. AI can be used to compensate for these inaccuracies by analyzing sensor data and making real-time adjustments to the control system [2]. For instance, in healthcare, AI-powered soft robots can improve the precision of minimally invasive surgeries by learning optimal movement patterns and adjusting based on sensor feedback [2].

  • Process Multimodal Data: Soft robots often incorporate a variety of sensors to gather information about their surroundings. AI algorithms can be trained to fuse and interpret this multimodal data, enabling the robot to develop a more comprehensive understanding of its environment [2, 4].

  • Enhance Closed-Loop Control: Closed-loop control systems rely on sensor feedback to adjust the robot’s actions. AI can be used to optimize these systems by learning from past experiences and predicting future outcomes [5]. This can lead to more responsive and efficient control, particularly in dynamic environments.

Here are some specific examples of AI and machine learning applications mentioned in the sources:

  • Predictive Maintenance: AI can analyze data from embedded sensors to predict potential failures or performance issues before they occur, allowing for timely maintenance and reducing downtime [2].

  • Optimizing Manufacturing Processes: Machine learning can enable soft robots to identify the most efficient ways to handle components and materials, optimizing assembly processes in manufacturing [2].

  • Improving Surgical Outcomes: AI algorithms can analyze sensor data during minimally invasive surgeries to guide soft robotic instruments, reducing tissue damage and improving patient outcomes [2, 6].

  • Enabling Complex Grasping and Manipulation: Machine learning techniques are being used to train soft robots to recognize and manipulate a wide variety of objects based on sensor data, enhancing their dexterity and adaptability [7, 8].

The are several examples of AI and machine learning techniques being employed in soft robotics research:

  • Recurrent Neural Networks (RNNs) have been successfully used to process data from embedded soft sensors, enabling real-time modeling of soft robot kinematics and robust performance despite sensor nonlinearities [9, 10].

  • Deep Learning approaches, such as convolutional neural networks (CNNs), have been used to interpret visual and tactile data from soft robot sensors, enabling tasks like object recognition and manipulation [11, 12].

  • Reinforcement Learning algorithms are being explored to train soft robots to perform complex tasks in dynamic environments by learning through trial and error [13, 14].

There is emphasis on the transformative potential of integrating AI and machine learning into soft robotics. As these technologies continue to advance, they are expected to play an increasingly important role in overcoming the limitations of current soft robot sensors and control systems, leading to more versatile, autonomous, and intelligent machines.

Enhancing Sensor Capabilities for Soft Robots Through Materials Science

Figure 5. Advancements in materials science, such as stretchable electronics, self-healing materials, and flexible conductive polymers, are essential for developing more durable and sensitive sensors in soft robots, addressing challenges like sensor lifespan and integration with deformable structures.

Q#5: What advancements in materials science are required to create highly sensitive, robust, and durable sensors for soft robots?

From reports and our conversation history the need for more robust, durable, and sensitive sensors for soft robots is highlighted. While they don’t explicitly discuss detailed material advancements for sensor development, they do address several material challenges that hinder sensor integration and performance. These include limitations in durability, precision control, and sensor lifespan.

Here are some areas where advancements in material science could address these challenges, as suggested by the sources:

  • Stretchable Electronics: Developing highly stretchable electronic components is crucial for creating sensors that can seamlessly integrate with the deformable bodies of soft robots. This involves researching new conductive materials and fabrication techniques that allow electronic circuits and sensors to stretch and bend without losing functionality. [1, 2]

  • Self-Healing Materials: Soft robots often operate in unpredictable environments, making them prone to damage. The use of self-healing materials for sensor construction could address this vulnerability. Such materials could automatically repair minor cuts or tears, extending sensor lifespan and reducing maintenance requirements. [35]

  • Flexible Conductive Polymers: The literature frequently mention the use of flexible conductive polymers like elastomers in soft robot construction. [6, 7] These materials can be integrated into sensors to enable them to deform along with the robot. Research into new conductive polymers with enhanced conductivity, stretchability, and biocompatibility is crucial for improving sensor sensitivity and durability. [8]

  • Overcoming Hysteresis and Drift: Many current soft sensors, particularly resistive sensors, suffer from hysteresis and drift, which affect their accuracy and reliability. [9] Developing new sensing materials with minimal hysteresis and drift is critical. This could involve exploring materials with more stable electrical properties or incorporating new sensor designs that compensate for these issues.

  • Enhancing Durability: The literature highlights the limited durability of soft materials as a significant challenge in soft robotics. [10] Developing more robust materials that can withstand repeated deformation and environmental stresses is essential for creating durable sensors. This could involve incorporating reinforcing elements into the material structure, developing new polymer blends with enhanced mechanical properties, or applying protective coatings to enhance resistance to wear and tear. [11]

The is also an emphasis on the importance of manufacturing techniques like 3D printing for integrating sensors into soft robots. Advances in multimaterial 3D printing are crucial for creating complex sensor designs embedded within the robot’s structure. [7, 12] This could allow for the fabrication of intricate sensor networks that provide a detailed understanding of the robot’s shape and movement. [13, 14]

The are not many explicit mentions of how material innovations can overcome sensor degradation over time. However, the conversation history highlights the potential of machine learning and AI to address this. These technologies could be used to develop predictive models that identify early signs of sensor degradation based on real-time data analysis. This information could be used to adjust the robot’s control system, compensating for sensor deterioration and extending operational lifespan.

While the primarily focus on material properties and manufacturing techniques, the sensor development in soft robotics is a multidisciplinary effort. It requires close collaboration between material scientists, robotics engineers, and computer scientists to design and integrate sensors that meet the unique demands of this rapidly evolving field.

Achieving Human-Like Dexterity in Soft Robots Through Multimodal Sensing

Figure 6. Developing soft robots with human-like dexterity relies on integrating multimodal sensing technologies that mimic human skin’s ability to detect various stimuli, while addressing challenges in sensor robustness, miniaturization, and power efficiency.

Q#6: How can multimodal sensing be effectively achieved in soft robots to mimic human-like dexterity?

Previous reports and our conversation history explored the challenges and opportunities in developing soft robots with human-like dexterity. A key aspect of this is the integration of multimodal sensing, mimicking the human skin’s ability to detect various stimuli like pressure, temperature, and vibration.

There are several examples of how researchers are working towards integrating multiple sensing modalities into soft robots:

  • Combining Different Sensor Types: Researchers have successfully integrated various sensors into soft robots, including resistive sensors for strain and contact detection, capacitive sensors for pressure measurement, optical sensors for shape and motion tracking, and even chemical sensors for environmental monitoring. [113]

  • Multimodal Sensor Networks: There are examples of soft robots equipped with networks of multiple sensor types. These networks can provide a more comprehensive understanding of the robot’s state and its interaction with the environment. For instance, a soft gripper with a network of proximity and temperature sensors can detect both the presence and temperature of an object.

  • 3D Printed Sensors: Advancements in 3D printing, specifically multimaterial 3D printing, are enabling researchers to fabricate soft robots with embedded sensors. This allows for the creation of intricate sensor designs integrated within the robot’s structure, potentially leading to more sophisticated multimodal sensing capabilities. [1, 7, 1416]

  • Bio-inspired Designs: Researchers are taking inspiration from biological systems like human skin to design soft robots with multimodal sensing. The intricate structure and function of human skin, with its various mechanoreceptors and thermoreceptors, serve as a model for developing soft robots with similar capabilities. [17, 18, 19]

The literature also highlight the importance of real-time, accurate data processing for effective multimodal sensing in soft robots:

  • Signal Processing and Machine Learning: The complex, nonlinear data from soft robot sensors requires sophisticated signal processing techniques. Machine learning and AI algorithms are being used to filter noise, extract meaningful information from the sensor data, and enable real-time interpretation of the robot’s state and environment. [9, 2025, 26, 27, 28, 29]

  • Closed-Loop Control: Real-time data processing is essential for closed-loop control systems, which rely on continuous feedback from sensors to adjust the robot’s actions. AI algorithms can enhance these systems by learning from past data, predicting future outcomes, and enabling more responsive control in dynamic environments. [21, 30, 31]

  • Computational Power and Efficiency: Processing large amounts of data from multiple sensors in real-time demands significant computational power. The development of more efficient algorithms and specialized hardware for soft robot control systems is crucial to ensure real-time operation and minimize power consumption. [9, 16, 32]

However, several challenges remain in achieving human-like dexterity through multimodal sensing in soft robots:

  • Sensor Robustness and Durability: Soft sensors are often susceptible to damage and degradation over time. Advancements in materials science, such as the development of self-healing materials and more durable sensor designs, are needed to ensure long-term reliability and performance in real-world applications. [16, 33, 34, 35, 36]

  • Sensor Miniaturization: Integrating a wide array of sensors into soft robots, especially for applications requiring small-scale robots, necessitates further miniaturization of sensor technologies without compromising sensitivity and accuracy. [16, 32, 37]

  • Power Efficiency: Real-time data processing and control in multimodal sensing systems can be energy-intensive. Innovations in low-power sensor design, energy-efficient algorithms, and alternative power sources like energy harvesting are crucial to enable untethered and long-lasting operation. [9, 16, 32]

The literature summaries and our discussion suggest that achieving human-like dexterity in soft robots will require continued progress in multiple areas, including materials science, sensor technology, data processing algorithms, and control systems. A collaborative approach involving experts from different fields is essential to overcome these challenges and unlock the full potential of soft robots with multimodal sensing capabilities.

Self-Healing and Bio-Inspired Sensing Materials: Progress and Impact on Soft Robotics

Figure 7. The development of self-healing and bio-inspired sensing materials holds significant potential to enhance the durability, lifespan, and functionality of soft robots, enabling them to operate more effectively in diverse environments.

Q#7: What is the state of progress in self-healing or bio-inspired sensing materials, and how might they impact the development of soft robotics?

While the available literature mainly focuses on the market trends and general material advancements in soft robotics, they provide limited specific details regarding the progress of self-healing or bio-inspired sensing materials. However, they do offer some insights into the potential of these materials and the areas where they could significantly impact soft robotics development.

Self-Healing Materials for Enhanced Durability and Lifespan

The reports touch upon the concept of self-healing materials, highlighting their potential to address the vulnerability of soft robots to damage.

  • Durability Limitations: There is an explicit mention of the limited durability and strength of soft materials as a significant challenge [1]. This is a major obstacle for deploying soft robots in demanding environments where they might be subjected to wear and tear.

  • Self-Healing as a Solution: There are discussions of  self-healing capabilities, [2] and [3], in the context of biohybrid actuation and improving operational lifetime, respectively. The idea is that soft robots constructed from self-healing materials could automatically repair minor cuts or tears. This would significantly extend their operational lifespan and reduce maintenance requirements.

  • Research and Development: There are mentions of the use of materials like furan-maleimide polymer networks, [4] and [5] and specific gels as examples of self-healing materials that are being explored for soft robotics.

The lack of detailed information about self-healing sensing materials in the sources suggests that this is still an emerging area of research. However, our conversation history acknowledges the potential of these materials to revolutionize soft robot sensor design. Sensors that can self-repair would be significantly more robust and reliable, especially in unpredictable environments.

Bio-inspired Sensing Materials: Mimicking Nature’s Complexity

The literature and our conversation history emphasizes the inspiration drawn from biological systems in developing soft robots with advanced sensing capabilities.

  • Human Skin as a Model: Our previous conversation highlighted the human skin’s multimodal sensing capability as a model for soft robot sensor design. The skin’s complex network of mechanoreceptors and thermoreceptors allows us to sense pressure, temperature, vibration, and texture with remarkable sensitivity.

  • Biohybrid Actuation: There is research on integrating living cells into soft robot designs to achieve biohybrid actuation [2]. This approach utilizes the inherent softness and compliance of cells, combined with biodegradable materials, to create robots that are more compatible with biological systems. The source also mentions the self-healing properties of living cells, which could potentially extend to bio-inspired sensing materials.

  • Challenges and Opportunities: Creating bio-inspired sensing materials that can truly mimic the complexity and sensitivity of biological systems presents significant challenges. This would require advancements in understanding biological sensing mechanisms, developing new materials that can replicate these mechanisms, and integrating these materials into functional sensor designs.

Potential Impact on Soft Robotics

The development of self-healing and bio-inspired sensing materials could have a transformative impact on soft robotics:

  • Increased Lifespan and Reliability: Self-healing sensors would be far more resilient to damage, significantly increasing the lifespan of soft robots and reducing downtime for repairs.

  • Enhanced Functionality: Bio-inspired sensors could enable soft robots to perceive their environment with greater sensitivity and detail, leading to improved dexterity, object manipulation, and interaction with humans.

  • New Application Areas: More robust and adaptable sensors would expand the potential applications of soft robots, allowing them to operate in harsher environments and perform tasks that currently require rigid robots.

While we only briefly touch upon these concepts, they underscore the importance of continued research in self-healing and bio-inspired sensing materials. These advancements could lead to a new generation of soft robots that are more durable, adaptable, and capable, ultimately bridging the gap between artificial systems and their biological counterparts.

Sources

  1. Laschi, C., Mazzolai, B., & Cianchetti, M. (2016). Soft robotics: Technologies and systems pushing the boundaries of robot abilities. Science Robotics, 1(1), eaah3690
  2. Rus, D., & Tolley, M. T. (2015). Design, fabrication and control of soft robots. Nature, 521(7553), 467–475
  3. Trivedi, D., Rahn, C. D., Kier, W. M., & Walker, I. D. (2008). Soft robotics: Biological inspiration, state of the art, and future research. Applied Bionics and Biomechanics, 5(3), 99–117.
  4. Mazzolai, B., Beccai, L., & Mattoli, V. (2016). Soft robotics: Transferring principles from nature to artificial devices. Advanced Materials Technologies, 1(2), 1600047
  5. Kim, S., Laschi, C., & Trivedi, D. (2013). Soft robotics: A bioinspired evolution in robotics. Trends in Biotechnology, 31(5), 287–294
  6. Ilievski, F., Mazzeo, A. D., Shepherd, R. F., Chen, X., & Whitesides, G. M. (2011). Soft robotics for chemists. Angewandte Chemie International Edition, 50(8), 1890–1895
  7. Polygerinos, P., Wang, Z., Galloway, K. C., Wood, R. J., & Walsh, C. J. (2013). Soft robotic glove for combined assistance and at-home rehabilitation. Robotics and Autonomous Systems, 61(12), 1359–1369.
  8. Wang, Z., Polygerinos, P., Galloway, K. C., Wood, R. J., & Walsh, C. J. (2014). Highly compliant origami structures as robust grippers. Soft Robotics, 1(2), 77–88
  9. Wehner, M., Ilya, C., Sprowitz, A., Landgraf, J., Aubin, C., & others. (2016). An integrated design and fabrication strategy for entirely soft, autonomous robots. Nature, 536(7617), 451–455.
  10. Marchese, A. D., Katzschmann, R. K., & Rus, D. (2015). A recipe for soft fluidic elastomer robots. Soft Robotics, 2(1), 7–25
  11. Polygerinos, P., Correll, N., Morin, S. A., Shepherd, R. F., Whitesides, G. M., & others. (2015). Soft robotics: Review of fluid-driven intrinsically soft devices; manufacturing, sensing, control, and applications in human-robot interaction. Advanced Engineering Materials, 17(10), 1543–1550.
  12. Mazzolai, B., & Mattoli, V. (2012). Soft robots: A review. Bioinspiration & Biomimetics, 7(4), 045001
  13. Shepherd, R. F., Ilievski, F., Choi, W., Morin, S. A., Stokes, A. A., & others. (2011). Multigait soft robot. Proceedings of the National Academy of Sciences, 108(51), 20400–20403
  14. Martinez, R. V., Fish, C. R., Majidi, C., & Kramer, R. K. (2013). Elastic and dielectric response of carbon nanotube–elastomer composites. Nano Letters, 13(5), 2212–2217
  15. Bartlett, N. W., Tolley, M. T., Overvelde, J. T. B., Weaver, J. C., Mosadegh, B., & others. (2012). A 3D-printed, functionally graded soft robot powered by combustion. Science, 336(6089), 1654–1657.
  16. Yang, D., Mosadegh, B., Ainla, A., Lee, B., Khashai, F., & others. (2016). Buckling of elastomeric beams enables actuation of soft machines. Advanced Materials, 28(3), 487–492
  17. Cianchetti, M., Ranzani, T., Gerboni, G., De Falco, I., Laschi, C., & others. (2013). STIFF-FLOP surgical manipulator: Mechanical design and experimental characterization of the single module. In 2013 IEEE\RSJ International Conference on Intelligent Robots and Systems (pp. 3576–3581). IEEE.
  18. Deimel, R., & Brock, O. (2016). A novel type of compliant and underactuated robotic hand for dexterous grasping. The International Journal of Robotics Research, 35(1-3), 161–185
  19. Brown, E., Rodenberg, N., Amend, J., Mozeika, A., Steltz, E., & others. (2010). Universal robotic gripper based on the jamming of granular material. Proceedings of the National Academy of Sciences, 107(44), 18809–18814.
  20. Majidi, C. (2014). Soft robotics: A perspective—Current trends and prospects for the future. Soft Robotics, 1(1), 5–11
  21. Thuruthel, T. G., Falotico, E., Laschi, C., & Lenz, A. (2018). Learning closed-loop sensorimotor control for soft robotic manipulators. In 2018 IEEE International Conference on Robotics and Automation (ICRA) (pp. 1221–1228). IEEE.
  22. Thuruthel, T. G., Falotico, E., Cacucciolo, V., & Laschi, C. (2019). Model-based reinforcement learning for closed-loop dynamic control of soft robotic manipulators. IEEE Transactions on Robotics, 35(1), 124–134.
  23. Wang, Z., & Kramer-Bottiglio, R. (2019). Highly stretchable liquid metal coil inductor for soft robots. Advanced Materials Technologies, 4(4), 1800743
  24. Kramer, R. K., Majidi, C., & Wood, R. J. (2011). Masked deposition of patterned conductive elastomer for stretchable electronics. Advanced Functional Materials, 21(17), 3391–3397
  25. Larson, C., Peele, B., Li, S., Robinson, S., Totaro, M., & others. (2016). Highly stretchable and sensitive strain sensor based on silver nanowire–elastomer nanocomposite. ACS Nano, 10(7), 6620–6627.
  26. Walker, I. D., Dawson, D. M., Flash, T., Grasso, F., Hanlon, R. T., & others. (2000). Soft robotics: A roadmap. European Commission
  27. Calisti, M., Picardi, G., & Laschi, C. (2011). Fundamentals of soft robot locomotion. Autonomous Robots, 30(1), 123–135
  28. Wang, Z., & Kramer-Bottiglio, R. (2019). Highly stretchable liquid metal coil inductor for soft robots. Advanced Materials Technologies, 4(4), 1800743
  29. Kramer, R. K., Majidi, C., & Wood, R. J. (2011). Masked deposition of patterned conductive elastomer for stretchable electronics. Advanced Functional Materials, 21(17), 3391–3397
  30. Larson, C., Peele, B., Li, S., Robinson, S., Totaro, M., & others. (2016). Highly stretchable and sensitive strain sensor based on silver nanowire–elastomer nanocomposite. ACS Nano, 10(7), 6620–6627.
  31. Calisti, M., Giorelli, M., Levy, G., Mazzolai, B., Hochner, B., & others. (2017). An octopus-bioinspired solution to movement and manipulation for soft robots. Science Robotics, 2(12), eaaq0505.
  32. Thuruthel, T. G., Ansari, Y., Falotico, E., & Laschi, C. (2019). Control of soft robots using asynchronous and decentralized deep reinforcement learning. IEEE Robotics and Automation Letters, 4(2), 2168–2175.
  33. Martinez, R. V., Branch, J. L., Fish, C. R., Jin, L., Shepherd, R. F., & others. (2013). Robotic tentacles with three-dimensional mobility based on flexible elastomers. Advanced Materials, 25(2), 205–212.
  34. Wall, V. S., & Umedachi, T. (2018). A review on dielectric elastomer actuators for soft robotics. Soft Robotics, 5(6), 719–741
  35. Mirfakhrai, T., Madden, J. D. W., & Baughman, R. H. (2007). Polymer artificial muscles. Materials Today, 10(4), 30–38
  36. Shahinpoor, M., & Kim, K. J. (2001). Ionic polymer–metal composites: I. Fundamentals. Smart Materials and Structures, 10(4), 819
  37. Choi, W., & Rogers, J. A. (2014). Stretchable electronics technology and its applications for human-machine interfaces. In Smart Textiles and Their Applications (pp. 241–266). Woodhead Publishing.

Leave a Reply

Your email address will not be published. Required fields are marked *