Researchers have introduced the Travelling Thief Problem with Drone (TTP-D), a new problem formulation that optimizes item selection, vehicle routing, and drone flight synchronization to maximize profit. The approach uses a mixed-integer linear program for small instances and develops metaheuristics and a deep reinforcement learning (DRL) policy for larger ones. A hybrid solver, combining the DRL policy with an annealing run, shows promise in recovering solution quality at a reduced computational cost. AI
IMPACT This research could lead to more efficient logistics and collection operations by optimizing the use of drones and ground vehicles.
RANK_REASON The cluster contains a research paper detailing a new problem formulation and solution methods.
Read on arXiv cs.NE (Neural & Evolutionary) →
- annealing
- arXiv
- deep reinforcement learning
- Travelling Thief Problem with Drone
- Hugging Face
- metaheuristics
AI-generated summary · Google Gemini · from 3 sources. How we write summaries →