Mobility by growth is a new paradigm in robotic systems design and their applications in the real world. Soft, tip-extending, or “growing”, robots have potential applications including inspection and navigation in disaster scenarios. However, due to their growing capability, such robots create unique challenges for intuitive human control. In this paper, a new flexible interface is proposed to intuitively map human bending commands into movements of the growing robot while providing shape information of the robot in order to improve situational awareness. Several command mappings are proposed, and a subjective study was conducted to assess the intuitiveness of the developed interface and mappings compared with other commercially available interfaces. The interfaces were evaluated using four metrics in two virtual task scenarios. The proposed interface with shape mapping performed better than the other interfaces, especially when the vine robot rolls over unintentionally during complex tasks.

Our research utilized deep learning to enhance the control of a 3 Degrees of Freedom biped robot leg. We created a dynamic model based on a detailed joint angles and actuator torques dataset. This model was then integrated into a Model Predictive Control (MPC) framework, allowing for precise trajectory tracking without the need for traditional analytical dynamic models. By incorporating specific constraints within the MPC, we met operational and safety standards. The experimental results demonstrate the effectiveness of deep learning models in improving robotic control, leading to precise trajectory tracking and suggesting potential for further integration of deep learning into robotic system control. This approach not only outperforms traditional control methods in accuracy and efficiency but also opens the way for new research in robotics, highlighting the potential of utilizing deep learning models in predictive control techniques.

Recently, flexible robots are growing in importance owing to their merits over rigid robots in maneuverability and safety, which equips them to work in unstructured environments, such as occur in medical applications. However, motion planning and control of flexible manipulators is challenging due to their compliance behavior and system uncertainties. Thus, this letter presents an Imitation-based Motion Planning (IbMP) approach, along with dynamic impedance control for learning, planning, and trajectory tracking of a two-section soft continuum robot in a dynamic environment. Point-to-point motion demonstrations, including the robot's tip position and orientation are intuitively provided by a motion capture system (OptiTrack V120-trio) and a similar kinematic flexible interface. Additionally, a singularity-free dynamic model based on Lagrangian formulation and Taylor expansion series of a two-section continuum robot is derived while planning for robot motions. The simulation results show that the IbMP approach, along with the dynamic impedance control, generalizes the robot's motion by varying the initial and goal of the robot's tip pose while avoiding static and dynamic obstacles, and moving back to the desired track after disturbances. Finally, the stability and performance of the IbMP algorithm against input disturbances are assessed using a Monte Carlo approach that can guide the selection of gain values.