Advances in Autonomous Systems and Biological Network Control

The field of autonomous systems and biological network control is rapidly advancing with the development of innovative methods and technologies. Recent research has focused on improving the safety and efficiency of autonomous vehicles, with a particular emphasis on scenario generation and prediction. Additionally, there have been significant advancements in the control of biological networks, including the use of graph neural networks and reinforcement learning to discover reprogramming strategies. Furthermore, researchers have explored the use of large language models and evolutionary strategies to control cellular dynamics and develop novel trajectory prediction heuristics. Notable papers in this area include: Seeking to Collide, which introduced an online, retrieval-augmented large language model framework for generating safety-critical driving scenarios. FalconWing presented an open-source, ultra-lightweight fixed-wing platform for autonomy research, demonstrating a purely vision-based control policy for autonomous landing. Graph Neural Network-Based Reinforcement Learning for Controlling Biological Networks proposed a novel control problem for Boolean network models and devised a computational framework to solve it. Giving Simulated Cells a Voice developed a functional pipeline that translates natural language prompts into spatial vector fields capable of directing simulated cellular collectives. Learn to Swim presented a Long Short-Term Memory network-based Fluid Experiment Data-Driven model for predicting unsteady, nonlinear hydrodynamic forces on an underwater quadruped robot. TrajEvo introduced a framework that leverages Large Language Models to automatically design trajectory prediction heuristics, outperforming previous heuristic methods and deep learning approaches.

Sources

Seeking to Collide: Online Safety-Critical Scenario Generation for Autonomous Driving with Retrieval Augmented Large Language Models

FalconWing: An Open-Source Platform for Ultra-Light Fixed-Wing Aircraft Research

Graph Neural Network-Based Reinforcement Learning for Controlling Biological Networks: The GATTACA Framework

Giving Simulated Cells a Voice: Evolving Prompt-to-Intervention Models for Cellular Control

Learn to Swim: Data-Driven LSTM Hydrodynamic Model for Quadruped Robot Gait Optimization

TrajEvo: Designing Trajectory Prediction Heuristics via LLM-driven Evolution

Built with on top of