Deep Reinforcement Learning for Time-Critical Wilderness Search And Rescue Using Drones

Traditional search and rescue methods in wilderness areas can be time-consuming and have limited coverage. Drones offer a faster and more flexible solution, but optimizing their search paths is crucial. This paper explores the use of deep reinforcement learning to create efficient search missions for drones in wilderness environments. Our approach leverages a priori data about the search area and the missing person in the form of a probability distribution map. This allows the deep reinforcement learning agent to learn optimal flight paths that maximize the probability of finding the missing person quickly. Experimental results show that our method achieves a significant improvement in search times compared to traditional coverage planning and search planning algorithms. In one comparison, deep reinforcement learning is found to outperform other algorithms by over $160\%$, a difference that can mean life or death in real-world search operations. Additionally, unlike previous work, our approach incorporates a continuous action space enabled by cubature, allowing for more nuanced flight patterns.

A Novel Methodology for Autonomous Planetary Exploration Using Multi-Robot Teams

One of the fundamental limiting factors in planetary exploration is the autonomous capabilities of planetary exploration rovers. This study proposes a novel methodology for trustworthy autonomous multi-robot teams which incorporates data from multiple sources (HiRISE orbiter imaging, probability distribution maps, and on-board rover sensors) to find efficient exploration routes in Jezero crater. A map is generated, consisting of a 3D terrain model, traversability analysis, and probability distribution map of points of scientific interest. A three-stage mission planner generates an efficient route, which maximises the accumulated probability of identifying points of interest. A 4D RRT* algorithm is used to determine smooth, flat paths, and prioritised planning is used to coordinate a safe set of paths. The above methodology is shown to coordinate safe and efficient rover paths, which ensure the rovers remain within their nominal pitch and roll limits throughout operation.

Enhancing Reinforcement Learning in Sensor Fusion: A Comparative Analysis of Cubature and Sampling-based Integration Methods for Rover Search Planning

This study investigates the computational speed and accuracy of two numerical integration methods, cubature and sampling-based, for integrating an integrand over a 2D polygon. Using a group of rovers searching the Martian surface with a limited sensor footprint as a test bed, the relative error and computational time are compared as the area was subdivided to improve accuracy in the sampling-based approach. The results show that the sampling-based approach exhibits a $14.75\%$ deviation in relative error compared to cubature when it matches the computational performance at $100\%$. Furthermore, achieving a relative error below $1\%$ necessitates a $10000\%$ increase in relative time to calculate due to the $\mathcal{O}(N^2)$ complexity of the sampling-based method. It is concluded that for enhancing reinforcement learning capabilities and other high iteration algorithms, the cubature method is preferred over the sampling-based method.