Evolutionary Algorithm Development: Created an evolutionary algorithm utilizing mutation and crossover techniques, enabling significant improvements in soft robots’ walking capabilities.
Custom Physics Engine: Engineered a physics engine from scratch to realistically simulate and evaluate the dynamic movements of evolving soft robots, ensuring accurate performance modeling.
Cloud-Based Computational Resources: Integrated Google Cloud Platform to scale computational resources, optimizing algorithm efficiency and allowing for faster evolution iterations.
Speed Milestone Achievement: Reached a walking speed of 0.6 m/s for evolved soft robots over 100,000 iterations, demonstrating the success of the algorithm and the precision of the physics engine.