Beyond Inverted Pendulums: Task-optimal Simple Models of Legged Locomotion
2301.02075
0
0
🤔
Abstract
Reduced-order models (ROM) are popular in online motion planning due to their simplicity. A good ROM for control captures critical task-relevant aspects of the full dynamics while remaining low dimensional. However, planning within the reduced-order space unavoidably constrains the full model, and hence we sacrifice the full potential of the robot. In the community of legged locomotion, this has lead to a search for better model extensions, but many of these extensions require human intuition, and there has not existed a principled way of evaluating the model performance and discovering new models. In this work, we propose a model optimization algorithm that automatically synthesizes reduced-order models, optimal with respect to a user-specified distribution of tasks and corresponding cost functions. To demonstrate our work, we optimized models for a bipedal robot Cassie. We show in simulation that the optimal ROM reduces the cost of Cassie's joint torques by up to 23% and increases its walking speed by up to 54%. We also show hardware result that the real robot walks on flat ground with 10% lower torque cost. All videos and code can be found at https://sites.google.com/view/ymchen/research/optimal-rom.
Create account to get full access
Overview
- Reduced-order models (ROMs) are popular for online motion planning due to their simplicity
- ROMs aim to capture critical task-relevant aspects of full dynamics while remaining low-dimensional
- Planning within the reduced-order space constrains the full model, sacrificing the robot's full potential
- This paper proposes an algorithm to automatically synthesize optimal ROMs for a given distribution of tasks and cost functions
- The authors demonstrate their approach on a bipedal robot Cassie, showing significant improvements in joint torque costs and walking speed
Plain English Explanation
Robots are complex machines with many moving parts, and it can be computationally intensive to plan their movements in real-time. Reduced-order models (ROMs) are simplified representations of the robot's full dynamics that are easier to work with. However, using a ROM means you can't take full advantage of the robot's capabilities.
This paper introduces a new way to create ROMs that are optimized for a specific set of tasks and goals. The researchers developed an algorithm that can automatically generate ROMs that balance simplicity and performance. They tested this approach on a bipedal robot called Cassie, and found that the optimized ROM reduced the robot's joint torque costs by up to 23% and increased its walking speed by up to 54% compared to a standard ROM.
The key idea is to treat the process of creating the ROM as an optimization problem. The algorithm searches for the ROM that best matches the robot's true dynamics for the specific tasks the robot needs to perform, as defined by the user. This allows the ROM to capture the most important aspects of the robot's behavior, without being overly complex.
Technical Explanation
The authors propose an optimization-based approach to automatically synthesize reduced-order models (ROMs) that are optimal with respect to a user-specified distribution of tasks and corresponding cost functions. In contrast to manually designing ROM extensions, this method provides a principled way to discover new model structures that balance simplicity and performance.
The core of their approach is a bilevel optimization problem. The upper-level optimization searches for the optimal ROM parameters, while the lower-level optimization solves trajectory optimization problems using the current ROM. The authors formulate the lower-level problem as a sequence of optimal control problems, which are solved using differential dynamic programming.
To demonstrate their method, the authors apply it to the bipedal robot Cassie. They define a distribution of walking tasks with different speeds and directions, and optimize the ROM accordingly. In simulation, the optimized ROM reduces Cassie's joint torque costs by up to 23% and increases its walking speed by up to 54% compared to a standard ROM. Hardware experiments also show a 10% reduction in torque costs on flat ground.
Critical Analysis
The authors acknowledge that their approach requires carefully defining the task distribution and cost functions, which can be challenging in practice. Additionally, the optimization process is computationally expensive, which may limit its real-time applicability. Further research could investigate ways to make the optimization more efficient, or to learn the task distribution from data rather than specifying it manually.
Another potential limitation is that the optimized ROM may only perform well within the specific task distribution used during training. Deploying the robot in novel scenarios could lead to suboptimal performance. Addressing this generalization challenge is an important area for future work.
That said, the authors' core idea of automating the process of ROM design is compelling. By framing it as an optimization problem, the method has the potential to discover model structures that outperform human-designed extensions, as demonstrated by the improvements on the Cassie robot. This principled approach to model reduction could benefit a wide range of robotic systems beyond legged locomotion.
Conclusion
This paper presents a novel algorithm for automatically synthesizing reduced-order models (ROMs) that are optimized for a given distribution of tasks and cost functions. By formulating ROM design as an optimization problem, the method can discover model structures that balance simplicity and performance in a principled way.
The authors demonstrate the effectiveness of their approach on a bipedal robot, Cassie, showing significant improvements in joint torque costs and walking speed compared to a standard ROM. While the method has some limitations in terms of computational complexity and generalization, it represents an important step towards more systematic model reduction techniques for robotic motion planning.
Ultimately, this work highlights the value of automated model discovery, which could unlock new capabilities for a wide range of robotic systems, control, and planning applications. As the field of task and motion planning continues to advance, techniques like the one presented in this paper will become increasingly important for enabling robots to operate effectively in the real world.
This summary was produced with help from an AI and may contain inaccuracies - check out the links to read the original source documents!
Related Papers
📈
Soft Synergies: Model Order Reduction of Hybrid Soft-Rigid Robots via Optimal Strain Parameterization
Abdulaziz Y. Alkayas, Anup Teejo Mathew, Daniel Feliu-Talegon, Ping Deng, Thomas George Thuruthel, Federico Renda
0
0
Soft robots offer remarkable adaptability and safety advantages over rigid robots, but modeling their complex, nonlinear dynamics remains challenging. Strain-based models have recently emerged as a promising candidate to describe such systems, however, they tend to be high-dimensional and time consuming. This paper presents a novel model order reduction approach for soft and hybrid robots by combining strain-based modeling with Proper Orthogonal Decomposition (POD). The method identifies optimal coupled strain basis functions -or mechanical synergies- from simulation data, enabling the description of soft robot configurations with a minimal number of generalized coordinates. The reduced order model (ROM) achieves substantial dimensionality reduction while preserving accuracy. Rigorous testing demonstrates the interpolation and extrapolation capabilities of the ROM for soft manipulators under static and dynamic conditions. The approach is further validated on a snake-like hyper-redundant rigid manipulator and a closed-chain system with soft and rigid components, illustrating its broad applicability. Finally, the approach is leveraged for shape estimation of a real six-actuator soft manipulator using only two position markers, showcasing its practical utility. This POD-based ROM offers significant computational speed-ups, paving the way for real-time simulation and control of complex soft and hybrid robots.
5/22/2024
Harnessing Natural Oscillations for High-Speed, Efficient Asymmetrical Locomotion in Quadrupedal Robots
Jing Cheng, Yasser G. Alqaham, Zhenyu Gan
0
0
This study explores the dynamics of asymmetrical bounding gaits in quadrupedal robots, focusing on the integration of torso pitching and hip motion to enhance speed and stability. Traditional control strategies often enforce a fixed posture, minimizing natural body movements to simplify the control problem. However, this approach may overlook the inherent dynamical advantages found in natural locomotion. By considering the robot as two interconnected segments, we concentrate on stance leg motion while allowing passive torso oscillation, drawing inspiration from natural dynamics and underactuated robotics principles. Our control scheme employs Linear Inverted Pendulum (LIP) and Spring-Loaded Inverted Pendulum (SLIP) models to govern front and rear leg movements independently. This approach has been validated through extensive simulations and hardware experiments, demonstrating successful high-speed locomotion with top speeds nearing 4 m/s and reduced ground reaction forces, indicating a more efficient gait. Furthermore, unlike conventional methods, our strategy leverages natural torso oscillations to aid leg circulation and stride length, aligning robot dynamics more closely with biological counterparts. Our findings suggest that embracing the natural dynamics of quadrupedal movement, particularly in asymmetrical gaits like bounding, can lead to more stable, efficient, and high-speed robotic locomotion. This investigation lays the groundwork for future studies on versatile and dynamic quadrupedal gaits and their potential applications in scenarios demanding rapid and effective locomotion.
5/29/2024
Adaptive Gait Modeling and Optimization for Principally Kinematic Systems
Siming Deng, Noah J. Cowan, Brian A. Bittner
0
0
Robotic adaptation to unanticipated operating conditions is crucial to achieving persistence and robustness in complex real world settings. For a wide range of cutting-edge robotic systems, such as micro- and nano-scale robots, soft robots, medical robots, and bio-hybrid robots, it is infeasible to anticipate the operating environment a priori due to complexities that arise from numerous factors including imprecision in manufacturing, chemo-mechanical forces, and poorly understood contact mechanics. Drawing inspiration from data-driven modeling, geometric mechanics (or gauge theory), and adaptive control, we employ an adaptive system identification framework and demonstrate its efficacy in enhancing the performance of principally kinematic locomotors (those governed by Rayleigh dissipation or zero momentum conservation). We showcase the capability of the adaptive model to efficiently accommodate varying terrains and iteratively modified behaviors within a behavior optimization framework. This provides both the ability to improve fundamental behaviors and perform motion tracking to precision. Notably, we are capable of optimizing the gaits of the Purcell swimmer using approximately 10 cycles per link, which for the nine-link Purcell swimmer provides a factor of ten improvement in optimization speed over the state of the art. Beyond simply a computational speed up, this ten-fold improvement may enable this method to be successfully deployed for in-situ behavior refinement, injury recovery, and terrain adaptation, particularly in domains where simulations provide poor guides for the real world.
4/19/2024
🛸
Task-Driven Computational Framework for Simultaneously Optimizing Design and Mounted Pose of Modular Reconfigurable Manipulators
Maolin Lei, Edoardo Romiti, Arturo Laurenz, Nikos G. Tsagarakis
0
0
Modular reconfigurable manipulators enable quick adaptation and versatility to address different application environments and tailor to the specific requirements of the tasks. Task performance significantly depends on the manipulator's mounted pose and morphology design, therefore posing the need of methodologies for selecting suitable modular robot configurations and mounted pose that can address the specific task requirements and required performance. Morphological changes in modular robots can be derived through a discrete optimization process involving the selective addition or removal of modules. In contrast, the adjustment of the mounted pose operates within a continuous space, allowing for smooth and precise alterations in both orientation and position. This work introduces a computational framework that simultaneously optimizes modular manipulators' mounted pose and morphology. The core of the work is that we design a mapping function that textit{implicitly} captures the morphological state of manipulators in the continuous space. This transformation function unifies the optimization of mounted pose and morphology within a continuous space. Furthermore, our optimization framework incorporates a array of performance metrics, such as minimum joint effort and maximum manipulability, and considerations for trajectory execution error and physical and safety constraints. To highlight our method's benefits, we compare it with previous methods that framed such problem as a combinatorial optimization problem and demonstrate its practicality in selecting the modular robot configuration for executing a drilling task with the CONCERT modular robotic platform.
5/6/2024