Field and Service Robotics

In this article, we first introduce a new research model proposed by the Science Council of Japan in 1999[1][2]. The model was proposed based on how research is carried out and how the research is integrated with our society as a culture. We consider what the Field and Service Robotics is, and introduce the Service RT Systems as a type of the field and service robots. Several examples illustrate the relation between the RT Systems and the Robotics Research.

After the debut of SDR-3X, SDR-4X made a stage appearance in 2002. It remains a small humanoid type robot and is expanding its capabilities of adaptability in home environment. Some of the new key technologies have been developed for SDR-4X. One is a Real-time Integrated Adaptive Motion Control. Another technology is a motion creating software system called “SDR Motion Creator” which allows to create and develop SDR’s attractive motion performances. In addition, speech synthesis and singing voice production are also developed for enhancement of entertainment applications. A cappella chorus performance and high-tempo dance performance are introduced as the attractive applications.

Autonomous mobile robots navigating in the real world are facing major challenges. They are expected to adapt best to different environments, interact with them and cope with sensor noise and incomplete information. Some important competencies of mobile robots, including locomotion, environment representation and navigation are discussed in this paper. Two research examples of our Lab demonstrate the application of this concepts: Innovative wheel-based locomotion concepts for rough terrain and a family of 11 tour guide robots with 5 month of operational experience.

In 1999 the German Federal Government launched six major strategic collaborative research projects on Human Technology Interaction, which involved 102 research partners and a funding volume of 82 million. The results of these projects were expected to allow people to control technical systems multimodally by using natural forms of interaction such as speech, gestures, facial expressions, touch and visualization methods and to apply such systems for the most varied purposes in their private and working environments. The ambitious research goals were achieved with prototypes for real-world applications. Research activities have resulted in 116 patent applications, 56 spin-off products and 13 spinoff companies as well as 860 scientific publications.

The paper discusses a new hybrid navigation strategy for mobile robots operating in indoor environment using the Information Assistant (IA) system and the Optical Pointer (OP). For intelligent navigation, the robots need a static and global information describing a topological map such as positional relation from any starting position to any goal position for making a path plan as well as dynamic and local information including local map, obstacles, traffic information for navigation control. We propose a method for managing the information. The robot has only rough path information to the goal, and the IAs, which are small communication devices installed in the environment, manage real environment information, locally. The OP is used for guidance of a robot in the junctions such as crossing, which communicates with mobile robots through IA and indicates their target positions by means of a light projection from a laser pointer onto the ground. The mobile robot allows it and run after the laser light beacon and reaches the destination. The robot can navigate to the goal efficiently by using these systems.

This paper presents a new wall-following algorithm for reactive navigation. It is based upon constrained active contours in a repulsive potential field. The development of this algorithm was initiated by the need for robust and efficient software for the navigation of large underground autonomous vehicles.

In this paper, we present a method which allows pose stabilization of a car-like vehicle to a learnt location based on feature bearing angle and range discrepancies between the vehicle’s current view of the environment, and that at the learnt location. We then extend the technique to include obstacle avoidance. Simulations and experimental results using our outdoor mobile platform are presented.

An approach to motion planning among moving obstacles is presented, whereby obstacles are modeled as intelligent decision-making agents. The decision-making processes of the obstacles are assumed to be similar to that of the mobile robot. A probabilistic extension to the velocity obstacle approach is used as a means for navigation as well as modeling uncertainty about the moving obstacles’ decisions.

Current methods for off-road navigation using vehicle and terrain models to predict future vehicle response are limited by the accuracy of the models they use and can suffer if the world is unknown or if conditions change and the models become inaccurate. In this paper, an adaptive approach is presented that closes the loop around the vehicle predictions. This approach is applied to an autonomous vehicle driving through unknown terrain with varied vegetation. Features are extracted from range points from forward looking sensors. These features are used by a locally weighted learning module to predict the load-bearing surface, which is often hidden by vegetation. The true surface is then found when the vehicle drives over that area, and this feedback is used to improve the model. Results using real data show improved predictions of the load-bearing surface and successful adaptation to changing conditions.

This paper presents a novel method for incorporating unstructured, natural terrain information into the process of tracking of underwater vehicles. Terrain-aided navigation promises to revolutionise the ability of marine systems to track underwater bodies in deepwater applications. This work represents a crucial step in the development of underwater technologies capable of long-term, reliable deployment. A particle based estimator is used to incorporate observations of altitude into the estimation process using a priori map information. Results of the application of this technique to the tracking of a towed body and a ship operating in Sydney Harbour are shown.

In this paper, we investigate the use of high resolution aerial LADAR (LAser Detection And Ranging) data for autonomous mobile robot navigation in natural environments. The use of prior maps from aerial LADAR survey is considered for enhancing system performance in two areas. First, the prior maps are used for registration with the data from the robot in order to compute accurate localization in the map. Second, the prior maps are used for computing detailed traversability maps that are used for planning over long distances. Our objective is to assess the key issues in using such data and to report on a first batch of experiments in combining high-resolution aerial data and on-board sensing.

The mobile robot which traverses on a rough terrain, should have an ability to recognize the shape of the ground surface and to examine the traversability by itself. Even if the robot is remotely controlled by operator, such an ability is still very important to save the operator’s load and to keep safety. For this purpose, the robot measures the front ground, and generate the local elevation map which represents the area where the robot is going to pass, and examines that the wheels can pass through and the bottom surface of robot body dose not contact with ground. This paper reports a simple method of traversability test for the wheeled mobile robot and show an experimental system with some results.

This is a technical report on an outdoor navigation for autonomous mobile robot based on DGPS and odometry positioning data. The robot position is estimated by fusion of DGPS and odometry. However, DGPS measurement data suffer from multi-path or other effects near high building and high trees. Thus, it is necessary to pick up only reliable and accurate DGPS measurement data when the robot position is corrected by data fusion of DGPS and odometry. In this paper, the authors propose a selection method of erroneous DGPS measurement data and a rule of data fusion for robot position correction. Finally, the authors equip a mobile robot with the proposed correction method and the robot autonomously navigates in a campus walkway.

The presented contribution describes an approach to preprocessing and fusion of additional vehicle onboard sensors – the odometer and accelerometer, all targeted to serve as optional and temporary substitute for GPS-like navigation. The suggested solution explores a rule-based system for mutual substitutions and calibrations of the used sensors depending on actual conditions. The only usage of the GPS here stands in providing regular position calibrations and serves as a reference method for evaluation of the presented results. The presented solutions have been experimentally tested with real-world data as shown in the experimental part of the paper.

Keywords:

data fusion, vehicle locator, odometer, accelerometer, odometry corrections, inertial navigation, satellite navigation

This paper studies the use of wireless Ethernet (Wi-Fi) as a localization sensor for mobile robots. Wi-Fi-based localization relies on the existence of one or more Wi-Fi devices in the environment to act as beacons, and uses signal strength information from those beacons to localize the robot. Through the experiments described in this paper, we explore the general properties of Wi-Fi in indoor environments, and assess both the accuracy and utility of Wi-Fi-based localization.

In this paper, the use of medial axis in mobile robot localization is extended. We describe the extraction of robust medial axis from a local occupancy grid to form a global map of the environment. We do not require any special geometric primitives to be present, and we allow substantial changes in the appearance of the environment, as long as the overall structure stays intact. We then describe how to use such a map with Monte Carlo Localization. Two approaches are treated. The first one relies on extracting salient features (tripods) from the local grid. The second one makes use of salient points in the global map and works well, even if no useful medial axis can be extracted during localization due to many new objects in formerly free space. This work aims to develop an efficient navigation system that requires less assumptions about the environment and its dynamics than current systems do.

In this paper, we introduce a novel approach to multirobot coordination that works by simultaneously distributing task allocation, mission planning, and execution among members of a robot team. By combining traditional hierarchical task decomposition techniques with recent developments in market-based multirobot control, we obtain an efficient and robust distributed system capable of solving complex problems. Essentially, we have extended the TraderBots market-based architecture to include a mechanism that distributes tasks among robots at multiple levels of abstractions, represented as task trees. Results are presented for a simulated area reconnaissance scenario.

This paper addresses the problem of simultaneous localization and mapping (SLAM) for teams of collaborating vehicles where the communication bandwidth is limited. We present a novel SLAM algorithm that enables multiple vehicles to acquire a joint map, but which can cope with arbitrary latency and bandwidth limitations such as typically found in airborne vehicle applications. The key idea is to represent maps in information form (negative log-likelihood), and to selectively communicate subsets of the information tailored to the available communication resources. We show that our communication scheme preserves the consistency, which has important ramifications for data association problems. We also provide experimental results that illustrate the effectiveness of our approach in comparison with previous techniques.

Inherent dangers in mining operations motivate the use of robotic technology for addressing hazardous situations that prevent human access. In the context of this case study, we examine the application of a robotic tool for map verification and void profiling in abandoned limestone mines for analysis of cavity extent. To achieve this end, our device enables remote, highly accurate measurements of the subterranean voids to be acquired. In this paper we discuss the design of the robotic tool, demonstrate its application in void assessment for prevention and response to subsidence, and present results from a case study performed in the limestone mines of Kansas City, Kansas.

A prerequisite to the design of future Advanced Driver Assistance Systems for cars is a sensing system providing all the information required for high-level driving assistance tasks. In particular, target tracking is still challenging in urban traffic situations, because of the large number of rapidly maneuvering targets. The goal of this paper is to present an original way to perform target position and velocity estimation, based on the occupancy grid framework. The main interest of this method is to avoid the decision problem of classical multitarget tracking algorithms. Obtained occupancy grids are combined with danger estimation to perform an elementary task of obstacle avoidance with an electric car.

This paper presents a Bayesian approach to the problem of searching for a single lost target by a single autonomous sensor platform. The target may be static or mobile but not evading. Two candidate utility functions for the control solution are highlighted, namely the Mean Time to Detection, and the Cumulative Probability of Detection. The framework is implemented for an airborne vehicle looking for both a stationary and a drifting target at sea. Simulation results for different control solutions are investigated and compared to demonstrate the effectiveness of the method.

This paper presents analysis of traction mechanics and control of a lunar/planetary rover based on the models obtained from terramechanics. A case study has been conducted for a rover test bed to negotiate a slope of loose soil such as regolith that covers most of lunar surface. The tire traction force is modeled as a function of the vertical load and slip ratio of the wheel. Bekker’s terramechanic formulae are employed to derive an improved practical model that calculates net traction force, referred to as Drawbar Pull, with a reasonable precision. Experiments are carried out in two phases. First, the physical behavior of a wheel on loose soil is observed using a single-wheel test bed, then the empirical parameters of the tire and soil are identified. Second, the slope climbing capability is studied by using a rover test bed that has independently driven four wheels. The traction margin and slip margin are defined to be used in a traction control. In the slope experiment, it turned out that the climbing capability was saturated at 14 degrees due to the lack of enough driving torque in wheels. But theoretical investigation suggests that this is not the limitation of terrain trafficability and climbing capability can be improved by increased driving torque and proper load distribution.

Robotic ground vehicles are systems that use gravity and contact forces with the ground to perform motion. In this paper we will focus on n-wheeled vehicles able to perform motion with all the wheels maintaining contact at the same time. The main goal of this work is to establish the implication of the topological architecture of the vehicle mechanism on criteria such as climbing skills, robustness, weight, power consumption, and price. Tools will be provided to help the robot designer to understand the implications of important design parameters like the number of wheels, the vehicle mechanism, and the motorisation of joints on the above criteria. Two examples of innovative locomotion concepts for rough terrain are presented and discussed.

We proposed a new holonomic mobile mechanism which is capable of running over the step. This mechanism realizes omni-directional motion on flat floor and passes over non-flat ground in forward or backward direction. The vehicle equips seven omni-directional wheels with cylindrical free rollers and two passive body axis that provide to change the shape of the body on the rough terrain. This paper presents a method to control the wheels for passing over rough terrain with the stable posture. Our vehicle is required to keep synchronization among its wheels for climbing the step without slipping and blocking. Therefore, in this paper, an algorithm of synchronization among all wheels is proposed. The performance of our system is experimented by means of computer simulations and experiments using our prototype vehicle.

A simple sensor-based walking on rough terrains for legged robots using an acceleration sensor attached to the body is described. The algorithm is implemented to a developed proto-type robot with limb mechanism, which has six limbs that can be used for both locomotion and manipulation. The six limbs are arranged on the body radially to have uniform property in all directions. This symmetrical structure allows the robot to generate a gait trajectory for omnidirectional locomotion in a simple manner. The trajectory of the sensorbased walking is obtained by a small conversion of this simple trajectory. The proto-type robot walks on the uneven ground while adjusting the pose of the body to keep high stability margin. Finally, adequate footholds of supporting limbs are examined for manipulation tasks by two neighboring limbs of the robot.

This paper details the development of a machine learning system which uses the helicopter state and the actions of an instructing pilot to synthesise helicopter control modules online. Aggressive destabilisation/restabilisation sequences are used for training, such that a wide state-space envelope is covered during training. The performance of heading, roll, pitch, height and lateral velocity control learning is presented using our Xcell 60 experimental platform. The helicopter is demonstrated to be stabilised on all axes using the “learning from a pilot” technique. To our knowledge, this is the first time a “learning from a pilot” technique has been successfully applied to all axes.

We present a vision-based algorithm designed to enable an autonomous helicopter to land on a moving target. The helicopter is required to identify the target, track it, and land on it while the target is in motion. We use Hu’s moments of inertia for precise target recognition and a Kalman filter for target tracking. Based on the output of the tracker, a simple trajectory controller is implemented which (within the given constraints) ensures that the helicopter is able to land on the target. We present data collected from manual flights which validate our tracking algorithm.

This paper describes initial results for a laser-based aerial mapping system. Our approach applies a real-time laser scan matching algorithm to 2-D range data acquired by a remotely controlled helicopter. Results obtain for urban and natural terrain exhibit an unprecendented level of spatial detail in the resulting 3-D maps.

Applying low-cost sensors for the Guidance, Navigation and Control (GNC) of an autonomous Uninhibited Aerial Vehicle (UAV) is an extremely challenging area. This paper presents the real-time results of applying a low-cost Inertial Measurement Unit (IMU) and Global Positioning System (GPS) receiver for the GNC. The INS/GPS navigation loop provides continuous and reliable navigation solutions to the guidance and flight control loop for autonomous flight. With additional air data and engine thrust data, the guidance loop computes the guidance demands to follow way-point scenarios. The flight control loop generates actuator signals for the control surfaces and thrust vector. The whole GNC algorithm was implemented within an embedded flight control computer. The real-time flight test results show that the vehicle can perform the autonomous flight reliably even under high maneuvering scenarios.

This paper presents a compact Millimeter Wave (MMW) radar unit that has been developed to be used as a Range, Bearing and Elevation (RBE) sensor on the Brumby Mk III Unmanned Air Vehicles (UAV). The Brumby MkIII is the flight platform used in the Autonomous Navigation and Sensing Experimental Research (ANSER) project which is focused on the development and demonstration of the Decentralised Data Fusion (DDF) and Simultaneous Localisation and Map Building (SLAM) algorithms on multiple UAVs. In the airborne DDF and SLAM demonstrations with UAVs, it is essential to have a terrain sensor on board for the RBE measurements of the ground targets. The focus of this paper is the hardware and software components of the MMW radar as an RBE sensor in the context of the ANSER project.

This paper presents a kinematic and force analysis of a mobile robot built on the principle of parallel mechanisms. The robot consists of an upper plate connected to 3 legs, each equipped with an asynchronous driving unit. A kinematic model for the robot provides data for accurate position estimate, even in rough and slippery terrains where conventional odometry fails. The paper presents an analysis of the forces acting on the robot under various surface conditions and robot configurations. This analysis provides useful data to determine whether a specific motion can be completed given the limitations on stability, the geometry and friction of the surface, and the required motion direction. The paper presents simulation results that are verified by experiments using our prototype model.

In this paper, we propose a collision avoidance algorithm for two nonholonomic tracked mobile robots transporting a single object based on a function-allocation concept. In this algorithm, a leader robot is controlled manually. A follower robot estimates the desired trajectory of the leader along its own heading direction and generates the motion for transporting the object and for avoiding obstacles by using an omni-directional vision sensor. We experimentally implement the proposed algorithm in a tracked mobile robots system, and illustrate the validity of the proposed control algorithm.

Dynamic stability is an important issue for vehicles which move heavy loads, turn at speed, or operate on sloped terrain. In many cases, vehicles face more than one of these challenges simultaneously. This paper presents a methodology for deriving proximity to tipover for autonomous field robots, which must be productive, effective, and self reliant under such challenging circumstances. The technique is based on explicit modeling of mass articulations and determining the motion of the center of gravity, as well as the attitude, in an optimal estimation framework. Inertial sensing, articulation sensing, and terrain relative motion sensing are employed. The implementation of the approach on a commercial industrial lift truck is presented.

In this paper, design and integration of ROBHAZ-DT2 is presented which is a newly developed mobile manipulator system. It is designed to carry out military and civilian missions in various hazardous environments. In developing the integrated ROBHAZ-DT2 system, we have focused on two issues: 1) novel mechanism design for mobility and manipulation and 2) intuitive user interface for teleoperation. In mechanism design, a double tracks connected by a passive joint has been designed to achieve high speed and rugged mobility on uneven terrain. In addition, a six-dof foldable manipulator suitable for the mobile manipulation has been designed. Secondly, a new compact 6-dof haptic device has been developed for teleoperation of the ROBHAZ-DT2. This haptic device is specially designed for simultaneous control both of the mobile base and the manipulator of ROBHAZ-DT2. As a result of integration of RobhAZ-DT2 and the user interface unit including the haptic device, we could successfully demonstrate a typical EOD task requiring abilities of mobility and manipulation in outdoor environment.

This paper introduces the Autonomous Vehicle Project (AVP) at the ANU together with a discussion and an example of driver assistance systems. A set of necessary core competencies of such a system is identified and in particular a system for force-feedback in the steering wheel when departuring from a lane is presented. A system like this is likely to reduce accidents due to driver fatigue since unintentional lane changes become more difficult. The presented system utilises a robust lane tracker which is experimentally evaluated for the purpose of driver assistance.

In the paper we consider skilled tasks that are performed by a human-like robot in outdoor unstructured environment when working interactively with a human operator. As skilled tasks we understand tasks, which – like when humans perform them – are non-trivial and demand learning or training. Typically such tasks may also fail and require several trials to be successfully executed. Performing requires use of senses and development of a unique miniplan, which takes into account the present situation. The mini-plan, which may also be called “skill”, produces a sequence of unit operations needed for successful performing of the task. The paper discusses preliminary experiences on implementing and experimenting such skilled tasks with a mobile centaur-like service robot.

To evaluate a neural interface, which comprises integral microelectrodes for neural recording or stimulation used for the control of human prosthetic devices, such as artificial limbs, a proto-type artificial body is being prepared for implantation of the interface in a rat. As a result of the first trial, new linkage mechanisms were developed, and a robot equipped with the mechanism was able to perform various walks likes an animal, including sidestep at high speed by swinging its legs.

This paper presents methods for cooperative mapping of partially or totally unknown area with human and robotic explorers. Mapping is supported with online modelling of presence, which will create a common understanding of the environment for both humans and robots. The studied key methods are human navigation without ready installed beacons, human and robotic SLAM, cooperative localization and cooperative map/model building for common presence. Methods are developed, tested and integrated in a European Community research project called PeLoTe.

Human care and service demands an innovative robotic solution to make easier the everyday of elderly and disabled people in home and workplace environments. The EU project MATS has been developing a new concept of climbing robot for this type of service applications. The robot developed by University Carlos III of Madrid is a 5 DOF selfcontaining manipulator, that includes on-board all the control system. The main advantage of the robot is its light weight, about 11 kg for a 1.3 m reach. The robot is a symmetrical arm able to move between different points (Docking Stations) of the rooms and, if it is necessary, “jump” to (or from) the environment to the wheelchair. In this way the MATS robot should became a home companion and assistance for numerous persons.

We describe a mobile robot system, designed to assist residents of an retirement facility. This system is being developed to respond to an aging population and a predicted shortage of nursing professionals. In this paper, we discuss the task of finding and escorting people from place to place in the facility, a task containing uncertainty throughout the problem.

Planning algorithms that model uncertainty well such as Partially Observable Markov Decision Processes (POMDPs) do not scale tractably to real world problems such as the health care domain. We demonstrate an algorithm for representing real world POMDP problems compactly, which allows us to find good policies in reasonable amounts of time. We show that our algorithm is able to find moving people in close to optimal time, where the optimal policy starts with knowledge of the person’s location.

This paper describes a personal service robot developed for assisting a user in his/her daily life. One of the important aspects of such robots is the user-friendliness in communication; especially, the easiness of user’s assistance to a robot is important in making the robot perform various kinds of tasks. Our robot has the following three features: (1) interactive object recognition, (2) robust speech recognition, and (3) easy teaching of mobile manipulation. The robot is applied to the task of fetching a can from a distant refrigerator.

This paper describes our continued work of a unique robotics project, Comprehensive Access to Printed Materials (CAPM), within the context of libraries. As libraries provide a growing array of digital library services and resources, they continue to acquire large quantities of printed material. This combined pressure of providing electronic and print-based resources and services has led to severe space constraints for many libraries, especially academic research libraries. Consequently, many libraries have built or plan to build off-site shelving facilities to accommodate printed materials. An autonomous mobile robotic library system has been developed to retrieve items from bookshelves and carry them to scanning stations located in the off-site shelving facility. This paper reviews the overall design of the robot system and control systems, and reports the new improvement in the accuracy of the robot performance; in particular, the pick-up process.

In this paper we report on the First International Contest for Cleaning Robots, which took place jointly with IROS 2002 in Lausanne, Switzerland. The event had two primary objectives. As an educational event with a significant fun factor it was supposed to attract the brainpower and activate the creativeness of students and young researchers for an application of service robotics, which has a significant economic potential. The cost for commercial cleaning services is estimated at around US$ 50 billion per year only in Europe. A fair contest, of course, required that all contestants had equal race conditions, This in turn required to have a well-define set up, which could be reproduced for every contest team and for any single run. With that, the second major objective of the event, which was to define a benchmark for robotic cleaning, was a natural byproduct of organizing a fair contest.

Since 1993 we have been working on the automation of dragline excavators, the largest earthmoving machines that exist. Recently we completed a large-scale experimental program where the automation system was used for production purposes over a two week period and moved over 200,000 tonnes of overburden. This is a landmark achievement in the history of automated excavation. In this paper we briefly describe the robotic system and how it works cooperatively with the machine operator. We then describe our methodology for gauging machine performance, analyze results from the production trial and comment on the effectiveness of the system that we have created.

This paper describes recent advances in the development of an integrated inertial guidance system for automation of the longwall coal mining process. Significant advances in longwall automation are being achieved through an industry sponsored project which targets productivity and safety benefits. Stabilised inertial navigation techniques are being successfully employed to accurately measure the three-dimensional path of the longwall shearer. This enabling technology represents a breakthrough in achieving practical and reliable automated face alignment. This paper also describes a specification for the interconnection of underground mining equipment based on the newly developed EtherNet/IP control and information protocol which ensures equipment compatibility across multi-vendor components of the automation system.

This paper describes the development of a new autonomous conveyor and bolting machine (ACBM) used for the rapid development of roadways in underground coal mines. The ACBM is a mobile platform fitted with four independent bolting rigs, bolt storage and delivery carousel, coal receiving hopper and through-conveyor for coal transport. The ACBM is designed to operate in concert with a standard continuous mining machine during the roadway development process to automatically insert roof and wall bolts for securing the roadway. This innovative machine offers significant benefits for increasing personnel safety and improving productivity. The paper describes the core sensing and processing technologies involved in realizing the level of automation required by the ACBM, which includes online roof monitoring, roadway profiling, navigation, and automatic control of drilling and bolting processes.

Mining operations depend on current, accurate maps of adjacent mine works to limit the risks of encroachment and breaching. Adjacent mines may be decades or centuries old with missing, inaccurate, or ambiguous maps. Dangers such as flooding, roof-fall, rotten support timbers, and poor ventilation preclude human entry to survey these spaces. Only robots may enter and directly observe these otherwise inaccessible underground voids, providing incontrovertible evidence of the mine’s existence and extent. This presents the configuration of a mobile mine mapping robot, Groundhog, and results from three deployments into coal mines.

For several years, our research group has been developing methods for automated modeling of 3D environments. In September, 2002, we were given the opportunity to demonstrate our mapping capability in an underground coal mine. The opportunity arose as a result of the Quecreek mine accident, in which an inaccurate map caused miners to breach an abandoned, water-filled mine, trapping them for several days. Our field test illustrates the feasibility and potential of high resolution three-dimensional (3D) mapping of an underground coal mine using a cart-mounted 3D laser scanner. This paper presents our experimental setup, the automatic 3D modeling method used, and the results of the field test. In addition, we address issues related to laser sensing in a coal mine environment.

There is a strong demand for efficient lifesaving techniques and devices in preparation for large-scale earthquakes. We focus on searching survivors and develop the rescue robot “Pneumatic-Drive Expandable Arm.” That is an elastic arm type robot driven by pneumatic pressure and has a camera on the head. That can travel stably in the rubble-strewn environment where electric power or wireless communication is not available.

This paper proposes a sufferers searching system using the group of robots to find sufferers at debris as quickly as possible in urban disaster. Five kind of new robots (Series UMRS-V) have developed as the searching robots and their hardware and software systems included the feature and mechanism, sensor system, data processing and control system have made clear. The human interface and simulator system have also developed to make the communication between robots and operator easy and to study the searching algorithm etc.

We develop a network of distributed mobile sensor systems as a solution to the emergency response problem. The mobile sensors are inside a building and they form a connected ad-hoc network. We discuss cooperative localization algorithms for these nodes. The sensors collect temperature data and run a distributed algorithm to assemble a temperature gradient. The mobile nodes are controlled to navigate using this temperature gradient. We also discuss how such networks can assist human users to find an exit. We have conducted an experiment at a facility used to train firefighters in order to understand the environment and to test component technology. Results from experiments at this facility as well as simulations are presented here.

A robot which could spray chemicals under grapevine trellis was developed and experimented. From the experimental results, it was observed that the robot system made precise spraying operation and its precise operation record possible. Based on the precise operations and records, an optimum management of chemicals could be expected, that is, necessary amount of chemicals would be sprayed only at necessary places for protection of environment and ecosystem. In addition, it was considered that this robot would be able to contribute the minimum input-maximum output production system by establishment of traceability system in grape production.

The problem of planning reference trajectories for agricultural machines is considered. A path planning algorithm to perform various kinds of farm-works is described. The case of convex fields is first considered. A direction of work being given, the algorithm determines the turning areas and selects a trajectory which guarantees the complete field coverage while minimizing overlapping. The method is extended to the case of fields with more complex shape including possibly obstacles. Simulations are proposed to illustrate the reasoning.