Plant-Inspired Soft Growing Robots: A Control Approach Using Nonlinear Model Predictive Techniques

Abstract

Soft growing robots, which mimic the biological growth of plants, have demonstrated excellent performance in navigating tight and distant environments due to their flexibility and extendable lengths of several tens of meters. However, controlling the position of the tip of these robots can be challenging due to the lack of precise methods for measuring the robots’ Cartesian position in their working environments. Moreover, classical control techniques are not suitable for these robots because they involve the irreversible addition of materials, which introduces process constraints. In this paper, we propose two optimization-based approaches, combining Moving Horizon Estimation (MHE) with Nonlinear Model Predictive Control (NMPC), to achieve superior performance in point stabilization, trajectory tracking, and obstacle avoidance for these robots. MHE is used to estimate the entire state of the robot, including its unknown Cartesian position, based on available configuration measurements. The proposed NMPC approach considers process constraints by relying on the estimated state to ensure optimal performance. We perform numerical simulations using the nonlinear kinematic model of a vine-like robot, one of the newly introduced plant-inspired growing robots, and achieve satisfactory results in terms of reduced computation times and tracking error.

Real-time deep learning-based model predictive control of a 3-DOF biped robot leg

“…Furthermore, a K-fold cross validation has been conducted to select the model architecture. The development of a Moving Horizon Estimation (MHE) 29 presents a promising avenue for future research, potentially mitigating the current assumption of full state observability that this work presupposes. Such advancements promise to further refine the precision and applicability of MPC in robotic leg control, contributing valuable insights into the integration of deep learning techniques within complex control systems.…”

Deep CNN-Based Static Modeling of Soft Robots Utilizing Absolute Nodal Coordinate Formulation

“…The reference trajectory is illustrated in Figure 9, highlighted in red and denoted as p re f ∈ R 12 with the corresponding coordinates indicating the tip orientation. The reference trajectory is computed based on the Constant Curvature Model (CCM) [17], which is widely adopted in defining the kinematics of a single-segment continuum robot. The reference rotation trajectory R ref and the reference positions x ref are chosen as…”

“…However, their naturally flexible and malleable nature makes modeling their dynamics a complex task [16]. Unlike robots with rigid components, continuum robots present both distinct challenges and potentials in the creation of controllers [17,18]. One of the key challenges in controlling continuum robots is achieving precise tip control, which is necessary for performing delicate tasks in complex environments [19,20].…”

Obstacle-Aware Navigation of Soft Growing Robots via Deep Reinforcement Learning

“…This irreversible nature of growth necessitates highly accurate and forward-thinking motion planning. In their recent work, [18] have pioneered the use of a Model Predictive Control (MPC) approach in the context of vine robot navigation, incorporating the kinematics of a vine robot as the predictive model, enabling advanced motion planning and obstacle avoidance. However, the development of an accurate kinematic model for vine-growing robots poses significant challenges.…”