Design, Control, and Actuation Optimization Strategies of Soft Robotic Joints: A Systematic Review Unveiling Novel Intelligent Material Frameworks and Adaptive Structure Innovations

Technical Summary: Design, Control, and Actuation Optimization Strategies of Soft Robotic Joints

Problem Statement
Traditional rigid robotic joints offer high precision and repeatability but pose significant safety risks in Human-Robot Interaction (HRI) and unstructured environments due to their inherent stiffness. They struggle with impact absorption and delicate handling, limiting their utility in applications such as prosthetics, surgery, and collaborative workspaces. While soft robotic joints inspired by biological systems (tendons, muscles, octopus arms) provide necessary compliance and safety, they face challenges regarding payload capacity, fabrication complexity, slower actuation speeds, fatigue, and a lack of standardized performance metrics. Furthermore, achieving multi-degree-of-freedom (multi-DoF) motion with real-time flexibility remains difficult due to control complexities.

Methodology
This study is a systematic review conducted in adherence to the PRISMA 2020 guidelines. The authors screened 2,414 articles published between January 2015 and December 2025 from seven databases (PubMed, ScienceDirect, MEDLINE, IEEE Xplore, Embase, Scopus, and Cochrane Library). The search utilized specific keywords related to soft robotic joints, smart materials, and actuation methods.

• Selection: After duplicate removal and screening of titles, abstracts, and full texts, 30 studies were selected for inclusion.
• Inclusion Criteria: Studies were required to feature physical prototypes, experimental validation with quantitative metrics (speed, force, torque), clear methodology, and reporting of limitations.
• Classification: The 30 selected studies were categorized into five primary groups based on dominant actuation and application:
  1. Pneumatically actuated soft robotic joints.
  2. Tendon- and cable-driven soft joints.
  3. Smart material-based soft joints (e.g., Shape Memory Alloys, Dielectric Elastomers).
  4. Hybrid soft-rigid/multi-material joint architectures.
  5. Wearable soft robotic joints for assistive and rehabilitative applications.
• Analysis Framework: The review employs a comparative performance framework analyzing four dimensions: Multi-DoF capability/Range of Motion (ROM), normalized torque/force output, stiffness modulation range, and durability (actuation cycles) with response dynamics.

Key Contributions and Results
The review synthesizes material-performance-control interdependencies and provides a benchmark comparison across the five categories.

1. Pneumatically Actuated Joints:

   • Performance: These designs demonstrate the highest force output, reaching up to 27 Nm torque and 200 N force. Stiffness can be increased 5–13 times through particle jamming or pressure modulation.
   • Limitations: Response speeds are generally lower (<1.2 Hz), and fatigue life is a concern compared to smart materials.
   • Examples: Biomimetic fingers using PDMS bellows, variable stiffness wrists using particle jamming, and origami-based actuators.

2. Cable and Tendon-Driven Joints:

   • Performance: These systems offer high precision and low backlash (<1.5° error). They support grip forces ranging from 2.68 N to 21.5 N with angular velocities up to 1.85 rad/s.
   • Innovation: Use of noncircular pulleys and harmonious cable mechanisms has been shown to reduce backlash by up to 40% and improve torque transmission efficiency by 30%.
   • Applications: Widely used in prosthetic hands (e.g., X-Limb) and wearable exosuits.

3. Smart Material-Based Joints:

   • Performance: Shape Memory Alloys (SMAs) and Dielectric Elastomers (DEs) exhibit high power density (up to 2250 W/kg) and exceptional durability (>100 million cycles). DEs can achieve area strains of 90% without prestretching.
   • Trade-offs: These materials often require specific electrical or thermal management (e.g., Joule heating for SMAs) and may have slower cooling phases.

4. Hybrid Soft-Rigid Architectures:

   • Performance: By combining soft and rigid components (e.g., Kevlar reinforcement, rigid frames with soft elastomers), these joints achieve omnidirectional motion (e.g., 30° bending, 44.2° twisting) and blocking torques up to 202 N·mm.
   • Efficiency: Hybrid solutions facilitate movement in all directions with efficiency gains of 20–50% compared to single-mode systems.

5. Wearable Assistive Joints:

   • Performance: Textile-based exosuits have demonstrated significant reductions in user muscle activity (43–63% reduction in bicep activity) while maintaining over 90% of the natural Range of Motion (ROM).
   • Weight: These systems are significantly lighter (0.6–1 kg) compared to traditional motor-driven exoskeletons (10–23 kg).

Significance and Claims
The paper claims that this systematic review establishes a foundational framework for understanding the trade-offs between different soft robotic joint strategies. It identifies that while no single technology dominates all metrics, specific architectures are superior for specific applications:

• Pneumatic and Hybrid systems are highlighted for intrinsic safety, high force, and adaptability in unstructured environments.
• Smart materials and Cable-driven systems are noted for superior speed, precision, and power density.
• Wearable systems are positioned as transformative for clinical rehabilitation due to their ability to reduce muscle fatigue and maintain natural kinematics.

The authors conclude that the field is shifting toward multimodal and hybrid systems to balance compliance with payload capacity. They propose that future advancements should prioritize:

1. AI-optimized topology optimization for multi-material hybrids.
2. Standardized constitutive models for variable stiffness.
3. Integration of embedded soft sensors and machine learning for real-time control.
4. Clinical validation of rehabilitation and prosthetic applications to bridge the gap between laboratory prototypes and real-world deployment.

The review asserts that these advancements mark a paradigm shift toward human-centered robotics, enabling safer and more adaptive interactions in healthcare, agriculture, and service sectors.