skip to main content

Title: Using First Principles for Deep Learning and Model-Based Control of Soft Robots
Model-based optimal control of soft robots may enable compliant, underdamped platforms to operate in a repeatable fashion and effectively accomplish tasks that are otherwise impossible for soft robots. Unfortunately, developing accurate analytical dynamic models for soft robots is time-consuming, difficult, and error-prone. Deep learning presents an alternative modeling approach that only requires a time history of system inputs and system states, which can be easily measured or estimated. However, fully relying on empirical or learned models involves collecting large amounts of representative data from a soft robot in order to model the complex state space–a task which may not be feasible in many situations. Furthermore, the exclusive use of empirical models for model-based control can be dangerous if the model does not generalize well. To address these challenges, we propose a hybrid modeling approach that combines machine learning methods with an existing first-principles model in order to improve overall performance for a sampling-based non-linear model predictive controller. We validate this approach on a soft robot platform and demonstrate that performance improves by 52% on average when employing the combined model.
; ; ; ;
Award ID(s):
Publication Date:
Journal Name:
Frontiers in Robotics and AI
Sponsoring Org:
National Science Foundation
More Like this
  1. Underactuation is a core challenge associated with controlling soft and continuum robots, which possess theoretically infinite degrees of freedom, but few actuators. However, m actuators may still be used to control a dynamic soft robot in an m-dimensional output task space. In this paper we develop a task-space control approach for planar continuum robots that is robust to modeling error and requires very little sensor information. The controller is based on a highly underactuated discrete rod mechanics model in maximal coordinates and does not require conversion to a classical robot dynamics model form. This promotes straightforward control design, implementation and efficiency. We perform input-output feedback linearization on this model, apply sliding mode control to increase robustness, and formulate an observer to estimate the full state from sparse output measurements. Simulation results show exact task-space reference tracking behavior can be achieved even in the presence of significant modeling error, inaccurate initial conditions, and output-only sensing.
  2. Learning policies in simulation is promising for reducing human effort when training robot controllers. This is especially true for soft robots that are more adaptive and safe but also more difficult to accurately model and control. The sim2real gap is the main barrier to successfully transfer policies from simulation to a real robot. System identification can be applied to reduce this gap but traditional identification methods require a lot of manual tuning. Data-driven alternatives can tune dynamical models directly from data but are often data hungry, which also incorporates human effort in collecting data. This work proposes a data-driven, end-to-end differentiable simulator focused on the exciting but challenging domain of tensegrity robots. To the best of the authors’ knowledge, this is the first differentiable physics engine for tensegrity robots that supports cable, contact, and actuation modeling. The aim is to develop a reasonably simplified, data-driven simulation, which can learn approximate dynamics with limited ground truth data. The dynamics must be accurate enough to generate policies that can be transferred back to the ground-truth system. As a first step in this direction, the current work demonstrates sim2sim transfer, where the unknown physical model of MuJoCo acts as a ground truth system.more »Two different tensegrity robots are used for evaluation and learning of locomotion policies, a 6-bar and a 3-bar tensegrity. The results indicate that only 0.25% of ground truth data are needed to train a policy that works on the ground truth system when the differentiable engine is used for training against training the policy directly on the ground truth system.« less
  3. Autonomous mobile robots (AMRs) have been widely utilized in industry to execute various on-board computer-vision applications including autonomous guidance, security patrol, object detection, and face recognition. Most of the applications executed by an AMR involve the analysis of camera images through trained machine learning models. Many research studies on machine learning focus either on performance without considering energy efficiency or on techniques such as pruning and compression to make the model more energy-efficient. However, most previous work do not study the root causes of energy inefficiency for the execution of those applications on AMRs. The computing stack on an AMR accounts for 33% of the total energy consumption and can thus highly impact the battery life of the robot. Because recharging an AMR may disrupt the application execution, it is important to efficiently utilize the available energy for maximized battery life. In this paper, we first analyze the breakdown of power dissipation for the execution of computer-vision applications on AMRs and discover three main root causes of energy inefficiency: uncoordinated access to sensor data, performance-oriented model inference execution, and uncoordinated execution of concurrent jobs. In order to fix these three inefficiencies, we propose E2M, an energy-efficient middleware software stack formore »autonomous mobile robots. First, E2M regulates the access of different processes to sensor data, e.g., camera frames, so that the amount of data actually captured by concurrently executing jobs can be minimized. Second, based on a predefined per-process performance metric (e.g., safety, accuracy) and desired target, E2M manipulates the process execution period to find the best energy-performance trade off. Third, E2M coordinates the execution of the concurrent processes to maximize the total contiguous sleep time of the computing hardware for maximized energy savings. We have implemented a prototype of E2M on a real-world AMR. Our experimental results show that, compared to several baselines, E2M leads to 24% energy savings for the computing platform, which translates into an extra 11.5% of battery time and 14 extra minutes of robot runtime, with a performance degradation lower than 7.9% for safety and 1.84% for accuracy.« less
  4. Autonomous educational social robots can be used to help promote literacy skills in young children. Such robots, which emulate the emotive, perceptual, and empathic abilities of human teachers, are capable of replicating some of the benefits of one-on-one tutoring from human teachers, in part by leveraging individual student’s behavior and task performance data to infer sophisticated models of their knowledge. These student models are then used to provide personalized educational experiences by, for example, determining the optimal sequencing of curricular material. In this paper, we introduce an integrated system for autonomously analyzing and assessing children’s speech and pronunciation in the context of an interactive word game between a social robot and a child. We present a novel game environment and its computational formulation, an integrated pipeline for capturing and analyzing children’s speech in real-time, and an autonomous robot that models children’s word pronunciation via Gaussian Process Regression (GPR), augmented with an Active Learning protocol that informs the robot’s behavior. We show that the system is capable of autonomously assessing children’s pronunciation ability, with ground truth determined by a post-experiment evaluation by human raters. We also compare phoneme- and word-level GPR models and discuss trade-offs of each approach in modeling children’smore »pronunciation. Finally, we describe and analyze a pipeline for automatic analysis of children’s speech and pronunciation, including an evaluation of Speech Ace as a tool for future development of autonomous, speech-based language tutors.« less
  5. Soft robots employ flexible and compliant materials to perform adaptive tasks and navigate uncertain environments. However, soft robots are often unable to achieve forces and precision on the order of rigid-bodied robots. In this paper, we propose a new class of mobile soft robots that can reversibly transition between compliant and stiff states without reconfiguration. The robot can passively conform or actively control its shape, stiffen in its current configuration to function as a rigid-bodied robot, then return to its flexible form. The robotic structure consists of passive granular material surrounded by an active membrane. The membrane is composed of interconnected robotic sub-units that can control the packing density of the granular material and exploit jamming behaviors by varying the length of the interconnecting cables. Each robotic sub-unit uses a differential drive system to achieve locomotion and self-reconfigurability. We present the robot design and perform a set of locomotion and object manipulation experiments to characterize the robot's performance in soft and rigid states. We also introduce a simulation framework in which we model the jamming soft robot design and study the scalability of this class of robots. The proposed concept demonstrates the properties of both soft and rigid robots, andmore »has the potential to bridge the gap between the two« less