Neuroevolution is a process of training neural networks (NN) through an evolutionary algorithm, usually to serve as a state-to-action mapping model in control or reinforcement learning-type problems. This paper builds on the Neuro Evolution of Augmented Topologies (NEAT) formalism that allows designing topology and weight evolving NNs. Fundamental advancements are made to the neuroevolution process to address premature stagnation and convergence issues, central among which is the incorporation of automated mechanisms to control the population diversity and average fitness improvement within the neuroevolution process. Insights into the performance and efficiency of the new algorithm is obtained by evaluating it on three benchmark problems from the Open AI platform and an Unmanned Aerial Vehicle (UAV) collision avoidance problem.
Majority of Artificial Neural Network (ANN) implementations in autonomous systems use a fixed/user-prescribed network topology, leading to sub-optimal performance and low portability. The existing neuro-evolution of augmenting topology or NEAT paradigm offers a powerful alternative by allowing the network topology and the connection weights to be simultaneously optimized through an evolutionary process. However, most NEAT implementations allow the consideration of only a single objective. There also persists the question of how to tractably introduce topological diversification that mitigates overfitting to training scenarios. To address these gaps, this paper develops a multi-objective neuro-evolution algorithm. While adopting the basic elements of NEAT, important modifications are made to the selection, speciation, and mutation processes. With the backdrop of small-robot path-planning applications, an experience-gain criterion is derived to encapsulate the amount of diverse local environment encountered by the system. This criterion facilitates the evolution of genes that support exploration, thereby seeking to generalize from a smaller set of mission scenarios than possible with performance maximization alone. The effectiveness of the single-objective (optimizing performance) and the multi-objective (optimizing performance and experience-gain) neuro-evolution approaches are evaluated on two different small-robot cases, with ANNs obtained by the multi-objective optimization observed to provide superior performance in unseen scenarios.Index Terms-Artificial neural network (ANN), experience gain, multi-objective, neuro-evolution of augmenting topologies (NEAT), unmanned ground vehicle (UGV)
This paper presents the conceptual design and fabrication/assembly of an autonomous solar powered small unmanned ground vehicle (UGV) platform for operation in outdoor environments. The contribution lies in the ability of the proposed design to offer uninterrupted operation in terms of endurance, to facilitate educational and research applications that are otherwise challenging to perform with a typical UGV (that needs significant downtime for recharging). A high incident area for solar PV panels is required to be able to support the complete energy needs of a ∼ 46 lb UGV (i.e., fully recharge the suitably sized battery powering the UGV). This makes it challenging to develop a stable platform that can carry solar panels much larger than the surface area of the platform itself (an aspect receiving minimal attention in other similar purpose platforms). To address this challenge, a novel umbrella-like folding mechanism is conceived, designed and successfully incorporated in the baseline prototype. This mechanism allows incorporating a remarkable ∼1 sq.m of incident solar PV with a net rated capacity of 200 W, one that remains folded to facilitate mobility, and can open/unfold to different extents for energy capture when needed. At the same time, the proposed design facilitates static and dynamic stability in spite of the significant solar PV incorporation. With the reference of the baseline prototype, an optimization approach is taken to develop a conceptual design of the next generation of this solar UGV. Specifically, the incident angle of the solar panels (enabled by the umbrella mechanism) at complete-open stage and the dimensions of the mechanism links and associated supports are separately optimized to respectively maximize the energy capture and the range of the UGV (assuming operation in Buffalo, NY), subject to stability and nominal velocity (of 2km/hr) constraints. The optimum design is found to provide an estimated range of 19.8 km/day.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.