Design with shape grammars and reinforcement learning

Shape grammars are a powerful and appealing formalism for automatic shape generation in computer-based design systems. This paper presents a proposal complementing the generative power of shape grammars with reinforcement learning techniques. We use simple (naive) shape grammars capable of generating a large variety of different designs. In order to generate those designs that comply with given design requirements, the grammar is subject to a process of machine learning using reinforcement learning techniques. Based on this method, we have developed a system for architectural design, aimed at generating two-dimensional layout schemes of single-family housing units. Using relatively simple grammar rules, we learn to generate schemes that satisfy a set of requirements stated in a design guideline. Obtained results are presented and discussed.     

Sindi: an intelligent assistant for highway design

We present a system to support highway engineers in the tasks of preliminary highway design. The system is based on the ideas of heuristic search and explicit preference setting. The system has been implemented and installed in a Spanish engineering firm.     

The autonomous mobile robot AURORA for greenhouse operation

AURORA has been conceived in order to substitute hard and unhealthy human work inside greenhouses by means of an autonomous mobile robot outfitted with appropriate sensors and operation devices. Emphasis has been put in the development of a new robotic platform specifically designed for greenhouse tasks, governed by a control architecture that supports both autonomous navigation and shared human control     

3D registration of laser range scenes by coincidence of coarse binary cubes

Scene registration of a pair of three-dimensional (3D) range images is a 6D optimization problem usually required in mobile robotics. This problem is different from object registration, since all scan directions and depths may contain relevant data, and because farther regions are sampled with lower densities. The paper proposes an efficient scene matching method based on the concept of coarse binary cubes. An integer objective function is defined as the number of coincident cubes between both scans. This is a metric of the degree of overlap that does not employ point distances. Its value is obtained without actually using any 3D grid data structure, with a computational complexity of order O(n), where n represents the number of laser points. This objective function is optimized with a globalized version of the downhill Simplex algorithm to avoid local minima. Experimental results are presented from indoor and outdoor environments with different degrees of structuring. The effect of cube size and the number of vertices on registration performance has been analyzed. Besides, experiments show that the proposed method achieves similar accuracy as iterative closest points (ICP) and normal distribution transform (NDT), while it improves both computation time and robustness against initial misalignments.     

Control architecture for mobile robot operation and navigation

This paper presents the navigation and operation system (NOS) for a multipurpose industrial autonomous mobile robot for both indoor and outdoor environments. This architecture supports task specification in terms of an event-driven state-based machine that provides high quality mission performance in uncertain environments. All processes in the NOS have been integrated in a distributed architecture designed to consider the real-time constraints of each control level of the system. Particular task models obtained from the system requirements specifications are integrated at the highest level of the architecture so that the rest of the levels remain unchanged for a wide range of industrial applications, such as transportation and operation with onboard devices.     

A comparison of multiobjective depth-first algorithms

Many real world problems involve several, usually conflicting, objectives. Multiobjective analysis deals with these problems locating trade-offs between different optimal solutions. Regarding graph search problems, several algorithms based on best-first and depth-first approaches have been proposed to return the set of all Pareto optimal solutions. This article presents a detailed comparison between two representatives of multiobjective depth-first algorithms, PIDMOA* and MO-DF-BnB. Both of them extend previous single-objective search algorithms with linear-space requirements to the multiobjective case. Experimental analyses on their time performance over tree-shaped search spaces are presented. The results clarify the fitness of both algorithms to parameters like the number or depth of goal nodes.     

Path recovery in frontier search for multiobjective shortest path problems

Frontier search is a best-first graph search technique that allows significant memory savings over previous best-first algorithms. The fundamental idea is to remove from memory already explored nodes, keeping only open nodes in the search frontier. However, once the goal node is reached, additional techniques are needed to recover the solution path. This paper describes and analyzes a path recovery procedure for frontier search applied to multiobjective shortest path problems. Differences with the scalar case are outlined, and performance is evaluated over a random problem set.     

Power Consumption Modeling of Skid-Steer Tracked Mobile Robots on Rigid Terrain

Power consumption is a key element in outdoor mobile robot autonomy. This issue is very relevant in skid-steer tracked vehicles on account of their large ground contact area. In this paper, the power losses due to dynamic friction have been modeled from two different perspectives: 1) the power drawn by the rigid terrain and 2) the power supplied by the motors. Comparison of both approaches has provided new insight on skid steering on hard flat terrains at walking speeds. Experimental power models, which also include traction resistance and other power losses, have been obtained for two different track widths over marble flooring and asphalt with Auriga- beta, which is a full-size mobile robot. To this end, various internal probes have been set at different points of the power stream. Furthermore, new energy implications for navigation of these kinds of vehicles have been deduced and tested.     

Mobile robot motion estimation by 2D scan matching with genetic and iterative closest point algorithms

The paper reports on mobile robot motion estimation based on matching points from successive two‐dimensional (2D) laser scans. This ego‐motion approach is well suited to unstructured and dynamic environments because it directly uses raw laser points rather than extracted features. We have analyzed the application of two methods that are very different in essence: (i) A 2D version of iterative closest point (ICP), which is widely used for surface registration; (ii) a genetic algorithm (GA), which is a novel approach for this kind of problem. Their performance in terms of real‐time applicability and accuracy has been compared in outdoor experiments with nonstop motion under diverse realistic navigation conditions. Based on this analysis, we propose a hybrid GA‐ICP algorithm that combines the best characteristics of these pure methods. The experiments have been carried out with the tracked mobile robot Auriga‐α and an on‐board 2D laser scanner. © 2006 Wiley Periodicals, Inc.