Siddharth Mysore

Bassel El Mabsout, Shahin Roozkhosh, Siddharth Mysore et al.

Fine-tuning simulation-trained RL agents with real-world data often degrades crucial behaviors due to limited or skewed data distributions. We argue that designer priorities exist not just in reward functions, but also in simulation design choices like task selection and state initialization. When adapting to real-world data, agents can experience catastrophic forgetting in important but underrepresented scenarios. We propose framing live-adaptation as a multi-objective optimization problem, where policy objectives must be satisfied both in simulation and reality. Our approach leverages critics from simulation as "anchors for design intent" (anchor critics). By jointly optimizing policies against both anchor critics and critics trained on real-world experience, our method enables adaptation while preserving prioritized behaviors from simulation. Evaluations demonstrate robust behavior retention in sim-to-sim benchmarks and a sim-to-real scenario with a racing quadrotor, allowing for power consumption reductions of up to 50% without control loss. We also contribute SwaNNFlight, an open-source firmware for enabling live adaptation on similar robotic platforms.

Aldair E. Gongora, Siddharth Mysore, Beichen Li et al.

Advancements in additive manufacturing have enabled design and fabrication of materials and structures not previously realizable. In particular, the design space of composite materials and structures has vastly expanded, and the resulting size and complexity has challenged traditional design methodologies, such as brute force exploration and one factor at a time (OFAT) exploration, to find optimum or tailored designs. To address this challenge, supervised machine learning approaches have emerged to model the design space using curated training data; however, the selection of the training data is often determined by the user. In this work, we develop and utilize a Reinforcement learning (RL)-based framework for the design of composite structures which avoids the need for user-selected training data. For a 5 $\times$ 5 composite design space comprised of soft and compliant blocks of constituent material, we find that using this approach, the model can be trained using 2.78% of the total design space consists of $2^{25}$ design possibilities. Additionally, the developed RL-based framework is capable of finding designs at a success rate exceeding 90%. The success of this approach motivates future learning frameworks to utilize RL for the design of composites and other material systems.

Siddharth Mysore, Bassel Mabsout, Renato Mancuso et al.

Actors and critics in actor-critic reinforcement learning algorithms are functionally separate, yet they often use the same network architectures. This case study explores the performance impact of network sizes when considering actor and critic architectures independently. By relaxing the assumption of architectural symmetry, it is often possible for smaller actors to achieve comparable policy performance to their symmetric counterparts. Our experiments show up to 99% reduction in the number of network weights with an average reduction of 77% over multiple actor-critic algorithms on 9 independent tasks. Given that reducing actor complexity results in a direct reduction of run-time inference cost, we believe configurations of actors and critics are aspects of actor-critic design that deserve to be considered independently, particularly in resource-constrained applications or when deploying multiple actors simultaneously.

Siddharth Mysore, Bassel Mabsout, Kate Saenko et al.

We focus on the problem of reliably training Reinforcement Learning (RL) models (agents) for stable low-level control in embedded systems and test our methods on a high-performance, custom-built quadrotor platform. A common but often under-studied problem in developing RL agents for continuous control is that the control policies developed are not always smooth. This lack of smoothness can be a major problem when learning controllers %intended for deployment on real hardware as it can result in control instability and hardware failure. Issues of noisy control are further accentuated when training RL agents in simulation due to simulators ultimately being imperfect representations of reality - what is known as the reality gap. To combat issues of instability in RL agents, we propose a systematic framework, `REinforcement-based transferable Agents through Learning' (RE+AL), for designing simulated training environments which preserve the quality of trained agents when transferred to real platforms. RE+AL is an evolution of the Neuroflight infrastructure detailed in technical reports prepared by members of our research group. Neuroflight is a state-of-the-art framework for training RL agents for low-level attitude control. RE+AL improves and completes Neuroflight by solving a number of important limitations that hindered the deployment of Neuroflight to real hardware. We benchmark RE+AL on the NF1 racing quadrotor developed as part of Neuroflight. We demonstrate that RE+AL significantly mitigates the previously observed issues of smoothness in RL agents. Additionally, RE+AL is shown to consistently train agents that are flight-capable and with minimal degradation in controller quality upon transfer. RE+AL agents also learn to perform better than a tuned PID controller, with better tracking errors, smoother control and reduced power consumption.

Siddharth Mysore, Bassel Mabsout, Renato Mancuso et al.

A critical problem with the practical utility of controllers trained with deep Reinforcement Learning (RL) is the notable lack of smoothness in the actions learned by the RL policies. This trend often presents itself in the form of control signal oscillation and can result in poor control, high power consumption, and undue system wear. We introduce Conditioning for Action Policy Smoothness (CAPS), an effective yet intuitive regularization on action policies, which offers consistent improvement in the smoothness of the learned state-to-action mappings of neural network controllers, reflected in the elimination of high-frequency components in the control signal. Tested on a real system, improvements in controller smoothness on a quadrotor drone resulted in an almost 80% reduction in power consumption while consistently training flight-worthy controllers. Project website: http://ai.bu.edu/caps