Using a Variational Autoencoder to Learn Valid Search Spaces of Safely Monitored Autonomous Robots for Last-Mile Delivery
This addresses the real-world challenge of safely scaling autonomous delivery robots for daily re-optimization, though it is incremental as it builds on existing hybrid machine-learning-optimization approaches.
The paper tackled the problem of optimizing autonomous robot timings for last-mile delivery to maximize deliveries while ensuring safe monitoring by limiting simultaneous robot operations, showing that the COIL method found valid solutions for all tested variations and optimized 10% faster than a genetic algorithm.
The use of autonomous robots for delivery of goods to customers is an exciting new way to provide a reliable and sustainable service. However, in the real world, autonomous robots still require human supervision for safety reasons. We tackle the realworld problem of optimizing autonomous robot timings to maximize deliveries, while ensuring that there are never too many robots running simultaneously so that they can be monitored safely. We assess the use of a recent hybrid machine-learningoptimization approach COIL (constrained optimization in learned latent space) and compare it with a baseline genetic algorithm for the purposes of exploring variations of this problem. We also investigate new methods for improving the speed and efficiency of COIL. We show that only COIL can find valid solutions where appropriate numbers of robots run simultaneously for all problem variations tested. We also show that when COIL has learned its latent representation, it can optimize 10% faster than the GA, making it a good choice for daily re-optimization of robots where delivery requests for each day are allocated to robots while maintaining safe numbers of robots running at once.