TetraRL: A Self-Adaptive Runtime for On-Device Deep Reinforcement Learning Systems
For developers of autonomous robotic systems, TetraRL provides a practical solution to manage multiple resource objectives simultaneously on resource-constrained hardware.
TetraRL is a self-adaptive runtime framework for on-device deep reinforcement learning that balances real-time performance, task reward, memory, and energy. It achieves competitive trade-offs under varying runtime preferences with negligible overhead on NVIDIA Jetson platforms.
Autonomous robotic systems, including autonomous vehicles, drones, and mobile robots, increasingly rely on on-device Deep Reinforcement Learning (DRL) to adapt to dynamic environments. Unlike cloud-based solutions, embedded DRL must perform training and inference directly on resource-constrained hardware while maintaining timely decision-making. This creates a fundamental challenge: balancing four tightly coupled objectives, real-time performance, task reward, memory utilization, and energy consumption. Optimizing these objectives independently often leads to suboptimal behavior, while conventional multi-objective methods may violate resource constraints and compromise reliability. This paper presents TetraRL, a self-adaptive runtime framework for tetra-objective on-device DRL. TetraRL formulates embedded DRL as a unified optimization problem over real-time, reward, RAM, and reserve (energy) objectives, and employs a preference-conditioned reinforcement learning controller to dynamically navigate the resulting trade-off space. The framework integrates a unified resource-management abstraction, hardware-aware DVFS control, and a runtime Override Layer for robust constraint enforcement. We implement TetraRL on NVIDIA Jetson AGX Orin and Orin Nano platforms and evaluate it across diverse DRL environments. Results show that TetraRL effectively balances all four objectives, achieves competitive trade-offs under varying runtime preferences, and incurs negligible overhead. Moreover, a single trained policy can support runtime-switchable optimization goals, providing a practical foundation for resource-aware and self-adaptive on-device DRL.