Feasible Static Workspace Optimization of Tendon Driven Continuum Robot based on Euclidean norm

This paper focuses on the optimal design of a tendon-driven continuum robot (TDCR) based on its feasible static workspace (FSW). The TDCR under consideration is a two-segment robot driven by eight tendons, with four tendon actuators per segment. Tendon forces are treated as design variables, while the feasible static workspace (FSW) serves as the optimization objective. To determine the robot’s feasible static workspace, a genetic algorithm optimization approach is employed to maximize a Euclidian norm of the TDCR’s tip position over the workspace. During the simulations, the robot is subjected to external loads, including torques and forces. The results demonstrate the effectiveness of the proposed method in identifying optimal tendon forces to maximize the feasible static workspace, even under the influence of external forces and torques.

💡 Research Summary

This paper presents a systematic approach to optimize the feasible static workspace (FSW) of a tendon‑driven continuum robot (TDCR) by treating tendon tensions as design variables and maximizing the Euclidean norm of the robot’s tip position. The robot under study consists of two segments, each actuated by four tendons arranged symmetrically around a compliant backbone, resulting in a total of eight tendons. The authors adopt a piecewise constant curvature (PCC) representation for the geometry, which simplifies the continuum shape to a series of circular arcs while preserving sufficient accuracy for workspace analysis.

A static model based on Newton‑Euler equilibrium is derived, incorporating tendon forces, gravity, friction, and external loads (forces and torques) applied at the tip. The tendon forces are decomposed into two vectors per tendon, and friction is modeled as a function of the normal force and a constant friction coefficient. Backbone bending and torsional stiffness are also included, yielding coupled nonlinear equations that describe the robot’s static configuration for any given set of tendon tensions.

The optimization problem is formulated as: maximize ‖p(T)‖, where p(T) is the tip position vector as a function of the tendon tension vector T. The tension variables are bounded by physically realistic lower and upper limits to avoid slack or rupture. The objective is expressed as the reciprocal of the Euclidean norm to fit the minimization framework of the genetic algorithm (GA).

A standard GA is employed: an initial population of random tension vectors is generated, fitness is evaluated as 1/‖p‖, and selection, crossover, and mutation operators evolve the population. Population sizes of 50, 100, and 200 were tested; the smallest size (50 individuals) achieved convergence within five generations, meeting a convergence criterion of 10⁻⁶ change in fitness. This demonstrates that the problem’s search space is well‑behaved and that the GA can find high‑quality solutions with modest computational effort.

Simulation results include a Monte‑Carlo style plot of 100 randomly sampled tension sets, showing the distribution of tip distances and a spline‑fitted envelope representing the reachable region. The GA‑optimized tension set significantly expands the envelope, even when a prescribed external force and torque are applied at the tip. Quantitatively, the optimized configuration yields tip distances up to 1.8 times larger than the average random configuration under the same external loading. The computational time is on the order of seconds, indicating suitability for rapid design iterations or even online adaptation in certain scenarios.

The authors discuss several limitations. The static model neglects dynamic effects such as inertia, damping, and vibration, which could be critical for high‑speed or impact‑prone tasks. Friction coefficients and backbone stiffness are assumed constant, whereas in practice they may vary with temperature, wear, or manufacturing tolerances. Consequently, future work should incorporate parameter identification, adaptive control, and multi‑objective optimization that balances workspace size, energy consumption, and structural robustness.

In conclusion, the paper introduces a clear and effective methodology for enlarging the static workspace of a tendon‑driven continuum robot by optimizing tendon tensions via a genetic algorithm. The approach is computationally efficient, robust to external loads, and directly applicable to medical minimally invasive devices or other applications where a large, reliable workspace within confined environments is essential.