New Insights for the Design of Bionic Robots: Adaptive Motion Adjustment Strategies During Feline Landings

Datao Xu, Huiyu Zhou, Xinyan Jiang, Shudong Li, Qiaolin Zhang, Julien S. Baker, Yaodong GuFrontiers in Veterinary Science 2022 · DOI

Abstract

Felines possess remarkable energy efficiency and flexibility, especially during high-impact activities like jumping and landing. This study explores the adaptive motion adjustment strategies employed by cats when landing from various heights (0.6m to 1.2m). By integrating experimental biomechanics, deep learning, and finite element analysis (FEA), researchers identified a critical "distal-to-proximal" shift in joint responsibility as impact forces increase. These findings offer a strategic blueprint for bionic robot design, advocating for intelligent joint coordination, targeted structural reinforcement, and posture-based stability.

Introduction: The Challenge of High-Impact Landing

While modern legged robots have achieved impressive locomotion on flat terrain, managing high-impact landings remains a significant engineering challenge. In nature, cats are the gold standard for landing safely from extreme heights. However, the specific coordination of their joints during these events has long been a "black box." This research aims to bridge the gap between biological excellence and robotic engineering by decoding how feline forelimbs—the primary impact buffers—adapt their kinematics and kinetics across different heights to prevent injury and maintain stability.

Methods: A Multi-Modal Pipeline

The study utilized a sophisticated multi-modal approach to analyze feline landing strategies:

  1. Experimental Biomechanics: 56 healthy adult cats were recorded jumping from four heights (0.6m, 0.8m, 1.0m, and 1.2m) onto a force plate. High-speed cameras captured the motion at 1000 fps.
  2. Inverse Kinetics: The feline forelimb was modeled as a three-link rigid body system (wrist, elbow, and shoulder) to calculate joint moments and forces.
  3. Machine Learning (DNN + LRP): A Deep Neural Network (DNN) classified landing heights based on joint angles and moments. Layer-wise Relevance Propagation (LRP) was then used to interpret the model, identifying which joints were most relevant to the landing strategy at specific heights.
  4. Finite Element Analysis (FEA): Using CT scans, a 3D model of the feline forelimb was constructed to simulate internal stresses at the moment of maximum impact.
Experimental setup and biomechanical modeling of feline landing, including the camera configuration (A), landing stages (B), and the multi-segment skeletal model (D, E).

Key Insights: Decoding the Feline Strategy

The analysis revealed several critical strategies that cats use to manage impact:

  • The Distal-to-Proximal Shift: This is the study's central finding. At lower heights (0.6m), the wrist (distal joint) is the primary contributor to impact absorption. As the height increases to 1.2m, the responsibility shifts significantly to the shoulder (proximal joint). This suggests that bionic robots require adaptive control logic that redistributes loads based on impact magnitude.
  • Stress Hotspots: FEA simulations showed that maximum stress concentrates at specific points, such as the styloid process of the ulna and the neck of the scapula. For engineers, this indicates that bionic legs should be specifically reinforced at these structural "bottlenecks" rather than just being made thicker overall.
  • The "Forward Lean" Strategy: At intermediate heights (0.8m), cats exhibit a specific posture—leaning forward with a low center of gravity—to minimize joint stress. This posture-based buffering is an energy-efficient way to improve stability without increasing motor power.

Key Results and Validation

  • Classification Accuracy: The DNN achieved a high accuracy rate (90.18% to 96.43%) in identifying landing patterns across different heights.
  • Joint Contribution: LRP analysis confirmed the wrist's high relevance (41.23%) at 0.6m and the shoulder's peak relevance (39.96%) at 1.2m.
  • FEA Validation: The FEA model was highly accurate, with only an 8% difference between predicted peak pressure (0.25 MPa) and experimental results (0.27 MPa). The shoulder joint consistently bore the highest maximum stress, confirming its role as the primary buffer for high-energy impacts.

Implications for Bionic Robotics

The research provides three primary recommendations for the design of bionic mechanical legs:

  1. Intelligent Coordination: Joint mechanisms should adjust according to the impact force, shifting the load from distal to proximal joints as force increases.
  2. Targeted Reinforcement: Materials around joint centers and structural narrow points should be strengthened to increase service life.
  3. Optimized Posture: Robots should maintain a low center of gravity and a forward-leaning posture to reduce machine wear and improve operational accuracy.

Limitations and Future Directions

While groundbreaking, the study has limitations. The FEA model treated bones as linear materials and did not fully account for the active damping of muscles and tendons. Future work will involve more diverse feline models and the physical implementation of these adaptive strategies in real-world bionic legs.

Summary updated

Abstract

Felines have significant advantages in terms of sports energy efficiency and flexibility compared with other animals, especially in terms of jumping and landing. The biomechanical characteristics of a feline (cat) landing from different heights can provide new insights into bionic robot design based on research results and the needs of bionic engineering. The purpose of this work was to investigate the adaptive motion adjustment strategy of the cat landing using a machine learning algorithm and finite element analysis (FEA). In a bionic robot, there are considerations in the design of the mechanical legs. (1) The coordination mechanism of each joint should be adjusted intelligently according to the force at the bottom of each mechanical leg. Specifically, with the increase in force at the bottom of the mechanical leg, the main joint bearing the impact load gradually shifts from the distal joint to the proximal joint; (2) the hardness of the materials located around the center of each joint of the bionic mechanical leg should be strengthened to increase service life; (3) the center of gravity of the robot should be lowered and the robot posture should be kept forward as far as possible to reduce machine wear and improve robot operational accuracy.
BibTeX
@inproceedings{xu2022new,
  author = {Datao Xu and Huiyu Zhou and Xinyan Jiang and Shudong Li and Qiaolin Zhang and Julien S. Baker and Yaodong Gu},
  title = {{New Insights for the Design of Bionic Robots: Adaptive Motion Adjustment Strategies During Feline Landings}},
  booktitle = {Frontiers in Veterinary Science},
  year = {2022},
  doi = {10.3389/fvets.2022.836043},
}
Figure