DEVELOPMENT OF THE AUTONOMOUS SURFACE CRAFT ACES

Transcription

1 DEVELOPMENT OF THE AUTONOMOUS SURFACE CRAFT ACES Justin E. Manley Massachusetts Institute of Technology Department of Ocean Engineering Sea Grant College Program Cambridge, MA Abstract - At the MIT Sea Grant College Program, Autonomous Surface Craft (ASCs) have been under development since An ASC is a small vessel outfitted with navigation and control systems which permit it to carry out functions autonomously. The first ASC developed was the ASC ARTEMIS, which was used to study command and control architectures, navigation systems, and basic data collection techniques. This vehicle successfully demonstrated the ability to operate autonomously and collect hydrographic data. ARTEMIS served well as a test platform but its small size made it unsuitable for coastal and open ocean research. Cruising speed, range, payload, and stability were all improved in the second generation ASC platform. The ASC ACES (Autonomous Coastal Exploration System) will provide better cruising speed, significantly more payload, longer mission endurance, and better seakeeping characteristics when compared to ARTEMIS. This was achieved within the primary design constraint of a 300 pound maximum weight chosen so that ACES could be easily deployed by a two member operations team. This paper describes the development of ACES. The challenge of designing a small lightweight platform with the required performance is explained and the solution implemented on ACES is described. Primary components of the design described here include the hull and structural members, propulsion and power systems, and steering systems. I. Introduction In an effort to reduce the costs of oceanographic research while increasing the amount of information available many new technologies have evolved. In recent years, the development of free roaming autonomous underwater vehicles (AUVs) and the concept of an Autonomous Ocean Sampling Network (AOSN) has introduced a new tool to the oceanographic community [l]. Building on the concelpt of autonomous ocean research the MIT Sea Grant Autonomous Surface Craft Laboratory has pioneered the development of free roaming robotic surface vessels to collect oceanographic data. Autonomous Surface Craft (ASCs) present several advantages over other oceanographic research systems. The Global Positioning System (GPS) has provided a high performance navigation aid, and commercially available satellite communication receivers allow for nearly instantaneous communication over much of the globe. This powerful naivgation and communications capability is an advantage available to surface craft. The capability of independent locomotion allows ASCs to provide a higher spatial distribution of data than moored buoys. Using these advantages the ASC Lab hopes to produce a valuable new research tool. A. The ASC ARTEMIS The first ASC produced at MIT Sea Grant was named ARTEMIS. This vessel is a 1/17 scale replica of a fishing trawler which was originally used for model basin testing. installation of an electric motor and a servo actuated rudder made the basic model into a platform capable of testing the navigation and control systems required by an ASC. Initial work focused on the development of control systems for the ASC. A microprocessor and digital compass were installed to provide rudimentary navigation and control functions. This configuration used a proportional-plusderivative (PD) control system to implement simple heading control. These first steps yielded an ASC with limited autonomy but provided a valuable proof of concept [2]. Follow on work added a Differential GPS (DGPS) receiver to enhance the navigation system of ARTEMIS. This facilitated the development of a heading constrained waypoint following controller based on Fuzzy Logic [3]. This provided the basic hardware and control systems for an ASC. To investigate the use of ASCs in data collection a depth sounder was added and ARTEMIS executed waypoint defined surveys to generate bathymetric maps of the Charles River in Cambridge, MA. The addition of a radio modem allowed these bathymetric maps to be generated in real time and provided human supervisory control of the ASC [4,5]. Fig. 1 is a picture of ARTEMIS in automated bathymetry configuration and Fig. 2 is a diagram of the system in operation /97/$ IEEE 827

2 ~~~ Fig. 1: The ASC ARTEMIS Differentially Corrected pounds. This limit permits the ASC to be transported, launched, and operated by a two member team. While manned vessels require launch ramps and significant resources, a lightweight ASC could be used in many confined or hard to reach bodies of water and in locations far from marinas or launch ramps. This would allow more coastal areas to be studied. In addition to increasing the speed, payload, and endurance of the next ASC, better seakeeping was required. This would provide the capability to operate in more exposed coastal waters. ARTEMIS could not operate in conditions more severe than those found on the Charles River. Even there, occasional waves of over one foot were experienced which were dangerous to the vehicle. A new design, which experienced less wave and wind induced motions than ARTEMIS, was called for. 11. The Development of the ASC ACES A. Hull and Structure Fig. 2: Automated Bathymetry Using ARTEMIS B. The Need for a New ASC At the conclusion of the automated bathymetry experiments with ARTEMIS several areas for improvement were identified in the platform characteristics. The specifications of ARTEMIS are given in Table 1. Length Beam Draft Maximum Displacement Endurance Speed - CruiselMaximum Table 1 : ARTEMIS Snecifications 54 inches 15 inches 8 inches 65 pounds 4 hours knots Three aspects identified for improvement were payload, endurance, and speed. A larger payload was desired because ARTEMIS (at 65 pounds) was fully laden and could not accommodate any new instruments or additional batteries. The speed and endurance of the next ASC were based on a desire to create a system as versatile and useful as a small manned vessel. A cruising speed of 5-7 knots and an endurance of 10 or more hours meet these goals. There is a desire to have the ability of launching an ASC from beach areas or areas with limited accessibility. To achieve this the ASC's maximum weight was limited to 300 The first task in the development of a new ASC was to design or select an appropriate hull form. One concept investigated was to modify a small kayak so that it was completely self-righting [6]. This provided a design which would have been robust enough for severe sea states. However, it did not provide enough roll stability for automated bathymetry which was selected as the first data collection mission for ACES. To provide enhanced roll stability and greater payload, a catamaran was selected as the best hullform for the new ASC. The wide beam and large waterplane area of catamarans reduces rolling motions and increases displacement without the significant drag penalty a similarly sized monohull would experience. This design also had the virtue of providing redundancy in the hull flotation. The failure of one hull would not result in a complete loss of buoyancy. The remaining hull could keep the ASC afloat long enough to be rescued. An investigation of small hulls which were commercially available led to the selection of the Hobie Float Cat line which offeredcatamaran hulls in 60 and 75 inch lengths. The maximum buoyancy of the 75 inch hull was 350 pounds. This was more than sufficient given that 300 pounds was established as the maximum gross weight for ACES. A 75 inch hull was obtained for use as the base for ACES. The frameworkincluded with the hull was designed to support a person on a small web seat. This was not an appropriate structure for the mounting of propulsion, steering, navigation, and control systems. A new structure consisting of a modular network of four longitudinal stringers attached to four cylindrical cross bars was designed for this purpose. structure allowed for flexible mounting of instruments and equipment. Some of the loading bays are designated for propulsion, steering, and vehicle control systems but the others can be configured to carry instruments or sensors a particular mission requires. 828

3 Another feature designed into the structure was a quick release mechanism. This allows the hulls to be removed from the structure so that the entire ASC can be broken down into small pieces for transport to an operation site. A further advantage of the simple mounting system is that different hulls can easily be installed onto the ACES structure allowing various hull configurations to be assembled for long range, high speed, or high sea statc operations. Fig. 3 shows the structure mounted between the hulls. interface allowed for easy integration into the autonomous system. The steering system chosen for ACES again focused on using simple and easily available components. A rudder mounting point was designed which would hold the skeg of a typical sailboard. This permits the use of most commercially available sailboard skegs which come in a great variety of materials and shapes. If a rudder breaks or does not perform satisfactorily under certain conditions it can easily be replaced. As with the throttle control, initial sea trials used a servo to actuate the rudder but were replaced by system using a stepper motor. C. The Advantages of ACES The final ACES platform fulfills the requirements identified for the successor to ARTEMIS. Table 2 presents the characteristics of ACES. Fig. 3: The ACES Main Structure B. Propulsion and Steering Once the hull and structure had been selected the issue of propulsion and power was resolved. Three options presented themselves; wind, electricity, or internal combustion. Wind was eliminated because autonomous control of sails was too complicated and a vessel propelled by sails would still require substantial electric power for instruments and actuators. A purely electrical system was ruled out by the weight limitation on ACES. An electric thruster and battery system which could provide 5-5' knot speeds for over 10 hours would be prohibitively heavy. Based on these concems, internal combustion was chosen to provide propulsion for ACES. A 3.3 hp gasoline engine was selected for installlation on ACES. The engine weighs 33 pounds and its fuel consumption rate requires approximately SO pounds of fuel to operate for 12 hours. For initial testing, electrical power for the computers, navigation, and control systems was provided entirely by batteries. A generator will be installed on the engine to recharge these batteries and make the fuel capacity of ACES the limiting factor on endurance. Limiting the total weight of the power and propulsion system to under a third of the total ASC weight is the primary advantage of using a gasoline engine. To actuate the engine throttle an electric servo wa:; used in initial sea trials. This allowed the basic ACES vehicle to be operated by radio control so that performance could be observed without using complicated computer control systems. The autonomous configuration replaces this servo with a stepper motor. The use of a commercially available stepper motor and motor controller which use an RS-232 Length Beam Draft Maximum Displacement 75 inches 51 inches 18 inches 350 pounds Endurance I hours Speed - CruiseiMaximum 5 I 10 knots Table 2 reveals that the endurance of ACES is at least three times that of ARTEMIS. The cruise and maximum speeds of ACES are 1.50 to over 300 percent faster than ARTEMIS. The maximum displacement of ACES is actually greater than the 300 pound limit specified. The weight of the hull, engine, and fuel is approximately 200 pounds. This permits a payload of 100 pounds and the hulls provide an additional 50 pounds of reserve buoyancy. These three criteria represent significant improvements over ARTEMIS without significantly increasing the complexity or size of the ASC. The speed and endurance of ACES make it comparable in performance to a small manned vessel. In terms of cost, ACES is also competitive with small manned vessels. By using off-the-shelfsystems and simple construction techniques the cost of the basic ACES platform was kept low. Prototype costs for the platform were approximately $7500. This includes several systems, such as the servo units used for radio control tests, which would not be needed on a production vehicle. The cost to manufacturea production ACES platform is estimated at $4500. The addition of a control, navigation and sensor suite like that used on ARTEMIS increases the cost to around $20,000. When compared to the costs of purchasing small boats, chartering research vessels, or deploying multiple fixed moorings the price of an ASC is competitive. The basic advantages of ACES, including speed, range, and payload were all quantified in the design stages. The goal of increased stability was identified in the design phase and computer aided naval architectural analysis indicated that 829

4 ACES would exhibit better stability than ARTEMIS. To verify the stability and the expected performance of ACES a series of sea trials were performed Sea Trials A. Radio Control Performance Assessment The first goal of initial sea trials was to obtain an impression of the performance of ACES. The speed, maneuverability and Bollard pull (a measure of the thrust available) were all parameters of interest The stability of the vehicle underway was also an important performancecriterion to observe. Sea trials were run from the MIT Sailing Pavilion in Cambridge, MA. These tests were performed under radio control. A standard marine radio control system was installed to operate the servos which controlled the throttle and rudder mechanisms. The range of this system was several hundred yards which allowed for a wide range of maneuvers. The vehicle s speed was measured by timing it as it traveled a known distance along the sailing pavilion dock. Idle speed was measured at 1.5 knots and one quarter throttle provided speeds of 3.0 knots. Additional speeds could not be recorded because ACES moved so fast that it was difficult to precisely follow the marked course and obtain accurate measurements. Maximum speed was estimated at 10 knots. The difficulty in recording higher speeds was a result of the impressive maneuverability of ACES. Even small rudder angles provided strong tuming responses. At low speeds ACES demonstrated a turning radius of approximately one boat length (6 feet). This exceptional tuming ability made it difficult for the operator to drive ACES along straight courses at higher speeds and prevented accurate speed measurements. Fig. 4 shows ACES performing a sharp turn. Fig. 4: ACES Demonstrates a Sharp Turn Measurements of the thrust provided by the engine yielded an explanation for the unexpected performance of the vehicle. A Bollard test was performed with a 20 pound scale. This proved to be insufficient as the thrust produced at full throttle easily exceeded20 pounds. This large amount of thrust, which was directed straight at the rudder, explained the speed and maneuverability of ACES. Unfortunately the thrust produced at high speed caused significant pitch changes. At higher speeds the stem of the vessel was pushed low in the water. This was not a problem at idle or lower speeds so it would not greatly influence any data collection performed at slow cruising speeds. The intended mission of bathymetric data collection would be significantly influenced by changes in vehicle pitch so correcting the high speed trim was identified as an important area for improvement in the ACES platform. While high speed operation caused some trim problems roll stability throughout the trials was good. Future tests will mount an inclinometer on ACES to measure the actual roll motions but visual observation of ACES underway revealed that it did not experience significant heeling motions. ARTEMIS rolled to angles estimated at 15 to 20 degrees but ACES rolled to maximum angles estimated at 10 degrees. This occurred only when the ASC crossed significant powerboat wakes which would have been dangerous to ARTEMIS. The radio controlled sea trials provided a valuable assessment of the performance of the ACES vehicle. In general ACES exceeded expectations. Speed, maneuverability, and thrust were all significantly better than was expected. Roll stability was also superior to ARTEMIS. The only negative feature of the vehicle s performancewas its tendency to pitch up while moving at high speed. This will be corrected by futuremodifications. B. Autonomous Systems Checkout After the radio control sea trials were performed, the rudder and throttle controls were upgraded to stepper motor based systems and the original control programs designed for servo systems were modified to accommodate the new actuators. The primary advantage of using these systems was that they provided feedbackabout the rudder angle or throttle position which allowed for more precise control over the vehicle. Once these enhancements were made, autonomous operations were planned. To gain experience with ACES under computer control the basic electronics package from ARTEMIS was installed. Short term battery and electronics housings were selected so that the preliminary autonomous tests would be easy to run. The microprocessor, DGPS, and various power and signal distribution boards were housed in a polycarbonate enclosure mounted to the foredeck. Batteries were mounted in two more enclosures on the port and starboard sides of the engine. A minimal battery set was installed to permit short check out missions. The field tests of ACES under autonomous control demonstrated that the basic control and navigation systems developed for ARTEMIS were also useful for a larger ASC. Three autonomous missions were run on the Charles River.

5 The procedure for each mission was quite simple. A waypoint defined course was downloaded to ACES. The engine was started by hand and the vehicle was launched. All three missions were performedat a constant speed. Fig. 5 shows ACES underway during one of the Autonomous missions. compartment for the batteries and as a measure to stabilize ACES. The additional weight will increase the roll stability and fins will be installed to actively or passively adjust the pitch of the hull at higher speeds. The keel will be used as a mounting point for sonars and other sensors. With the installation of a generator and the battery housing keel the only limiting factor on the endurance of ACES will be its fuel supply. The hulls of the vehicle will be modified to accommodate extra fuel. This will provide for long range missions of hour endurance. B. Electronics Upgrades Fig. 5: ACES Operating Autonomously The first mission was a simple checkout of the ASC systems. A single waypoint was identified and ACES was commanded to navigate to that point and shut down. The second mission added an additional waypoint which (defined an out and back course for ACES. This mission was also successful. The final mission was defined by three waypoints and defineda T shaped course. Again ACES hit all three waypoints and correctly executed its programmed mission. These missions demonstrated the validity of the control and navigation systems which had already been employed on ARTEMIS and the enhanced performanceof ACES. In early work with ARTEMIS the ASC would follow a relatively straight course but would slowly oscillate to either side of its intended heading as it progressed towards a waypoint. ACES, in contrast, followed courses which were remarkably straight and experienced little deviation from its defined course. This improvement may be attributable to the better stability and maneuverability of ACES. The autonomous trials were run on the Charles River but they proved to be a mild simulation of ocean conditions because they were run in a heavy rainstorm. The amount of water which covered the ASC was comparable to what it would experience if it was powering though substantial waves. While large wave actions were not experienced, ACES did endure a thorough soaking and some substantial powerboat wakes which would have posed a significant threat to ARTEMIS. A. Platform Enhancements IV. Future Plans To maximize the potential of ACES, several modifications to the mechanical systems are planned. A generator will be installed on the engine to recharge the operating batteries. In addition, a watertight cylindrical keel will be mounted between the hulls. This will serve as a The mechanical improvements of ACES will be complemented by upgrades to the electronic systems. The single computer will be replaced by a several networked microprocessors. This will allow various functions, including navigation, data collection, and control, to be distributed to multiple processors. This configuration will allow significant increases in the complexity of the vehicle. A new DGPS system with its own base station will be installed to provide enhanced navigation. This system will provide accuracies of 0.5 m in real time and 2 cm accuracywith post-processing of the position data. ACES will also be equipped with two scanning single beam sonars. One will provide obstacle avoidance information and the other will collect bathymetric data. Eventually the data collection sonar will be replaced by a multi-beam system to provide complete bottom coverage. C. Future Missions Once the mechanical and electrical improvements are complete ACES will undergo a period of extensive field testing. Experiments to develop obstacle avoidance and survey pattern behaviors will be carried out in the Atlantic Ocean off Gloucester, MA. These will be followed by a long endurance mission. ACES will cruise between Gloucester and Provincetown, MA in order to demonstrate its ability to operate for extended periods in the open ocean. These missions are planned for the summer of Further developments beyond that are planned and the ASC Lab has set a goal of sending an ASC across the Atlantic from the United States to Europe. D. Scientific Missions While the ASC Lab will continue to perform demonstration missions of increasing complexity, scientific applications of ASCs are planned as well. Initial scientific work will build on the early bathymetric survey experience with ARTEMIS. An ASC will be used to provide accurate and inexpensive bathymetric surveys. Applications for this vehicle include support of dredging operations and updating of outdated nautical charts. Additional missions planned for ASCs include monitoring of water quality in drinking water reservoirs, study of beach erosion in near coastal waters, and as a data link to submerged instruments. It is the goal of the 83 1

6 ASC Lab to develop ASCs to fulfill many important scientific and commercial roles. V. Conclusion The MIT Sea Grant Autonomous Surface Craft Lab has developed two small, easily deployed ASCs for use in ocean research. The ASC ARTEMIS was a testbed system which facilitated the development of navigation and control systems. ARTEMIS also pioneered the autonomous collection of data. ARTEMIS was followed by the development of the ASC ACES. Use of off-the-shelfkomponents and previously proven electronic systems allowed ACES to be developed quickly and inexpensively. A catamaran hull and gasoline engine were the major innovations in the ACES vehicle. Field tests demonstrated the performance of ACES mechanical systems. Speed, thrust, and stability, were all observed to be better than ARTEMIS during these trials. The final specifications of the ACES ASC are a cruising speed of 5-7 knots, a payload of at least 100 pounds, and an endurance of over 10 hours. Missions and improvements are planned which will turn ACES into a mature, robust, and useful scientific tool for the exploration and study ofthe oceans. Future goals of the ASC Lab are to develop ASCs capable of carrying out a wide range of missions and with ocean crossing performance. Autonomous Surface Craft represent a unique new tool with exciting potential for ocean research. [4] C.D. Rodriguez-Ortiz, Autonomous Bathymetry Using an Autonomous Surface Craft, Master s Thesis, MIT Department of Ocean Engineering. September, [5] T. Vaneck, J. Manley, C. Rodriguez, and M. Schmidt, Automated Bathymetry using an Autonomous Surface Craft, NAVIGATION, Joumal of the Institute of Navigation, [6] J. Manley, A Preliminary Design Study for an Autonomous Surface Craft Society of Naval Architects and Marine Engineers New England Section Student Paper, February, VI. Acknowledgments The author would like to acknowledge the support of his colleagues and fellow students at the MIT Sea Grant Program. Dr. Thomas Vancek, director ofthe ASC Lab, has been especially helpful in providing guidance on the mechanical design of ACES. The contribution of a Hobie Float Cat from the Hobie Cat Company is also gratefully acknowledged. The work described here is supported by the MIT Sea Grant College Program and the Department of Ocean Engineering. VII. References [l] T. Curtin, J. Bellingham, J. Catipovic, and D. Webb, Autonomous Ocean Sampling Networks, Oceanography, Vol. 6 No. 3, [2] J. Manley and M. Frey, Development and Operation of the Autonomous Surface Craft ARTEMIS. MIT Sea Grant Undergraduate Summer Research Program [3] T. Vaneck, Fuzzy Guidance Controller for an Autonomous Boat, IEEE Control Systems Magazine, April,