An official website of the United States government
Catastrophic forgetting continues to severely restrict the learnability of controllers suitable for multiple task environments. Efforts to combat catastrophic forgetting reported in the literature to date have focused on how control systems can be updated more rapidly, hastening their adjustment from good initial settings to new environments, or more circumspectly, suppressing their ability to overfit to any one environment. When using robots, the environment includes the robot's own body, its shape and material properties, and how its actuators and sensors are distributed along its mechanical structure. Here we demonstrate for the first time how one such design decision (sensor placement) can alter the landscape of the loss function itself, either expanding or shrinking the weight manifolds containing suitable controllers for each individual task, thus increasing or decreasing their probability of overlap across tasks, and thus reducing or inducing the potential for catastrophic forgetting.
more »
« less
Automated shapeshifting for function recovery in damaged robots
A robot's mechanical parts routinely wear out from normal functioning and can be lost to injury. For autonomous robots operating in isolated or hostile environments, repair from a human operator is often not possible. Thus, much work has sought to automate damage recovery in robots. However, every case reported in the literature to date has accepted the damaged mechanical structure as fixed, and focused on learning new ways to control it. Here we show for the first time a robot that automatically recovers from unexpected damage by deforming its resting mechanical structure without changing its control policy. We found that, especially in the case of "deep insult", such as removal of all four of the robot's legs, the damaged machine evolves shape changes that not only recover the original level of function (locomotion) as before, but can in fact surpass the original level of performance (speed). This suggests that shape change, instead of control readaptation, may be a better method to recover function after damage in some cases.
more »
« less
- Award ID(s):
- 1830870
- PAR ID:
- 10122109
- Date Published:
- Journal Name:
- Proceedings of Robotics: Science and Systems (2019)
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
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, and has the potential to bridge the gap between the twomore » « less
-
Integrating self-healing materials with structural health monitoring (SHM) represents a significant advancement in materials science and engineering. In applications such as aerospace, self-healing composites with SHM can detect and repair damage autonomously, enhancing safety and reducing maintenance time, which significantly improves structural durability by maintaining integrity and extending material service life. This research employs an experimental approach to validate the self-healing process of self-healing metal matrix composites reinforced with shape memory alloy fibers, by utilizing lead zirconate titanate (PZT) piezoelectric transducers mounted on the surface of the composite. A three-point bending test is used to induce damage in the metal matrix composites by applying a load at its midpoint while supporting it at both ends; then, self-healing is used to return the specimen to its original state. The ultrasonic signals from the PZT sensor were compared at three stages: pristine state, during bending (i.e., damage), and after the healing process, to evaluate the effectiveness of the healing process and health monitoring technique adopted for this study. Real-time monitoring using digital image correlation, laser profilometry, and mechanical testing was used to validate the damaged/healed signal state—demonstrated by improved recovered signal amplitude, reduced scatter energy, a notable decrease in damage indices root-mean-square deviation, normalized scatter energy, and stabilized wave propagation. The successful self-healing composite design and health monitoring results confirm the restoration of the specimen, which regained 84% of its initial shape deformation at 135 °C for 30 min (i.e., during the first step of healing). Complete shape restoration was achieved by raising the temperature to 150 °C for 90 min (during the second step of healing), recovering approximately 96% of the original flexural strength after healing.more » « less
-
Robots can influence people to accomplish their tasks more efficiently: autonomous cars can inch forward at an intersection to pass through, and tabletop manipulators can go for an object on the table first. However, a robot's ability to influence can also compromise the physical safety of nearby people if naively executed. In this work, we pose and solve a novel robust reach-avoid dynamic game which enables robots to be maximally influential, but only when a safety backup control exists. On the human side, we model the human's behavior as goal-driven but conditioned on the robot's plan, enabling us to capture influence. On the robot side, we solve the dynamic game in the joint physical and belief space, enabling the robot to reason about how its uncertainty in human behavior will evolve over time. We instantiate our method, called SLIDE (Safely Leveraging Influence in Dynamic Environments), in a high-dimensional (39-D) simulated human-robot collaborative manipulation task solved via offline game-theoretic reinforcement learning. We compare our approach to a robust baseline that treats the human as a worst-case adversary, a safety controller that does not explicitly reason about influence, and an energy-function-based safety shield. We find that SLIDE consistently enables the robot to leverage the influence it has on the human when it is safe to do so, ultimately allowing the robot to be less conservative while still ensuring a high safety rate during task execution. Project website: https://cmu-intentlab.github.io/safe-influence/more » « less
-
For decades, the field of biologically inspired robotics has leveraged insights from animal locomotion to improve the walking ability of legged robots. Recently, “biomimetic” robots have been developed to model how specific animals walk. By prioritizing biological accuracy to the target organism rather than the application of general principles from biology, these robots can be used to develop detailed biological hypotheses for animal experiments, ultimately improving our understanding of the biological control of legs while improving technical solutions. Much of this work involves biologically inspired walking controllers informed by the morphology and dynamics of the insect nervous system, which necessitate a robot with highly animal-like structure to prevent a brain-body mismatch. However, methods for codifying suitable fidelity in biomimetic robots currently vary, with limited generalizable methods for robot design. In this work, I outline a general framework for developing biomimetic robots that ensures kinematic and dynamic similarity between the robot and target animal. I then use this framework to develop and validate the robot Drosophibot II, a meso-scale robotic model of an adult fruit fly, Drosophila melanogaster. The resulting robot is novel for its close attention to the kinematics and dynamics of Drosophila, an increasingly important model of legged locomotion. Each leg’s proportions and degrees of freedom are modeled after Drosophila 3D pose estimation data. The predominant actuators for the robot are characterized to determine their inertial, elastic, and viscous properties and subsequently dynamically scale the robot's motions. I then use a developed program to automatically solve the inverse kinematics and inverse dynamics necessary for walking for the robot's structure and that of a to-scale model of the fly. By comparing the output of these models, I demonstrate that the robot and fly are kinematically and dynamically similar. The robot's electromechanical design is presented, then validated by having the robot’s walk forward, backward, and up an incline via open-loop straight line stepping with biologically inspired foot trajectories. Strain data from locations throughout the robot's legs is also recorded during these tests as an analog for mechanosensory feedback in a freely walking animal. Through these experiments, Drosophibot II demonstrates its utility for neuromechanical investigations by providing plausible neural data currently unobtainable in the animal.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.15607/RSS.2019
