PROJECT IDEA SUBMISSION STUDENT Team Contacts - 1 st person listed serves as the point of contact with Professor Jensen - Initial team size may be from 4 to 6 members (all members must agree to have their name included) NAME PHONE EMAIL Yashvin Vedanaparti 610-806-3879 yvedanap@purdue.edu Alex Eib 574-354-8068 eib@purdue.edu Sanket Jaiswal 317-270-7779 jaiswas@purdue.edu Michael Jung Note: Course policy prohibits projects involving weapons, or significant safety risk. PROJECT NAME Battery Swapping Robot PROJECT DESCRIPTION Our goal is to demonstrate the function of an autonomous robot such that it could replace its own battery. Autonomous robots come in many forms today, and the industry is rapidly growing. The limiting factor with these robots, however, is battery power. We want to develop a robot-dock system, in which an autonomous robot can navigate to a charging dock and have a battery replacement operation performed. This function would enable robots to operate independently with a negligible turnaround time compared to recharging. After discussing with several professors, we believe that this function could have application in various industries, and can expand the capabilities of autonomous robots. WHAT IS THE PROBLEM YOU ARE TRYING TO SOLVE? Many autonomous robots exist nowadays, from warehouse robots to vacuum cleaners
to drones. A problem with this category of robots, however, is that they usually run on a rechargeable battery. When the robot is running low, a human user has to physically change the battery, charge the robot, or in some cases, the robot is adept enough to find its own charging pad, but the user must still wait for the robot to charge up. The turnaround time between these methods reduces the productivity of these robots. If a mechanism allowed these robots to swap their own batteries, the turnaround time is greatly reduced and human dependency can also be eliminated. WHY IS IT A PROBLEM WORTH SOLVING? Autonomous robots are becoming more and more prevalent in our world and have greater productivity performing some tasks compared to humans. The limiting factor (charging), however, reduces their effectivity. If the function we hope to develop can be demonstrated, then this industry will generate more productive, independent robots. This is better for businesses and people that utilize autonomous robots. MARKET ANALYSIS: WHO ARE THE TARGET CUSTOMERS AND HOW WILL THEY BENEFIT FROM YOUR PROPOSED PRODUCT? The target customer will be companies that utilize autonomous robots. In particular, Amazon appropriates warehouse robots that retrieve orders throughout the facility. These are battery powered, and need to be charged every hour. They run on a rotating fleet based system where 5% of the robots are out of commission at any given time. This system does not maximize the potential of these robots. A battery swapping system would mean that ideally 100% of the robots are in commission at all times. It also would make battery maintenance easier, since a robot would not need to be decommissioned to fix a bad battery. Battery degradation limits a robot s lifespan to about 2 years. Making this function modular would change that. We would also consider irobot s Roomba. These autonomous robots can vacuum one s home and have an automatic self charge system. The solution that we propose, however, would enable these robots to quickly return to 100% battery, and not have to charge. The reduction in turnaround time would make the robot more efficient for its users. BENCHMARK RESEARCH: WHAT SOLUTIONS EXIST TODAY AND HOW IS YOUR PROPOSED PRODUCT SOLVING THE PROBLEM UNIQUELY AND BETTER? DESCRIBE RELEVANT PATENTS AND COMMERCIAL PRODUCTS INCLUDING THEIR FUNCTIONAL PERFORMANCE AND COST. The concept of autonomously replacing the battery has been considered by some big companies and startups. Tesla had acquired a patent for a battery swapping station and had briefly implemented it before closing the program and deciding to expand their supercharging network. This robotic system was augmented by humans and reportedly
had some technical issues. Another pursuit of this concept was achieved by the Israeli startup Airobotics, which developed a drone and station system that can autonomously swap batteries. That being said, this concept is not widespread in the autonomous robot industry. Current market solutions generally involves rotating fleets, where one fleet of robots is working and the other is charging. The problem with this, however, is rotating fleets involve having more robots than necessary. It also requires more space to charge a fleet of robots versus batteries. The solution we plan to implement, therefore, would be a cheaper alternative to fleet rotation and also take up less space. Furthermore, rechargeable batteries deteriorate over the course of a few years. When they are built in, the robot becomes less efficient over time. Our solution would make the battery packs modular, and thus dealing with this issue is as simple as putting in a new battery. DESIGN: DESCRIPTION OF YOUR PROPOSED SOLUTION(S). DESCRIBE THE FUNCTIONAL REQUIREMENTS FOR THE FINAL PRODUCT. INCLUDE PRELIMINARY SKETCHES OF THE INITIAL CONCEPT(S), AND EXPLAIN HOW THEY ARE INNOVATIVE. There will be two separate robots for the prototype. One is a mobile unit that can move around, and the other is the stationary charging dock. The mobile unit will be powered by a small processor (Raspberry Pi, Arduino, or myrio), which will control the positioning of the robot. It will use infrared sensors for obstacle detection, and navigate accordingly. It will also use these to locate the charging dock. A control system will be implemented (LabView if using myrio, or python if other) to align the mobile unit with the dock. The dock will operate a mechanical arm with two degrees of freedom. The arm will be used to remove the battery and install the new one. The location of the battery on the robot will be such that it makes removal quick. Below are rough schematics depicting the concept. The drawing on the left is of the charging station and the arm. The drawing on the right is of the mobile unit that would navigate to find the dock.
SYNTHESIS: WHAT ENGINEERING PRINCIPLES WILL BE USED ON THE PROJECT AND HOW? The project will rely a lot on the principles learned in Control Systems classes (ME 375, 475) and possibly machine design (ME 352, 452). There needs to be a closed feedback loop that will dictate the robot positioning relative to the charging dock. The mechanism on the dock used to replace the battery will likely involve a kinematic analysis and another feedback loop to ensure that the new battery is placed in the same position each time. Putting everything together will involve a good amount programming as well, and a sufficient understanding of electrical systems. ENGINEERING: EXPLAIN THE PHYSICS/ENGINEERING ASSOCIATED WITH THE PROBLEM. (Include Free-body diagrams or other sketches as necessary) The problem will involve developing precise control systems. The solution to the issue we tackle will have two components: an autonomous robot and a servicing charging dock. The first task would involve the robot finding the charging dock, which would be fixed, and aligning itself into the proper position. This process would involve an efficient feedback loop that allows the robot to detect the deviation from desired position and self adjust. Once it connects to the dock, the dock would perform the replacement operation. This would also require an additional control system. MANUFACTURING: EXPLAIN HOW THE PROTOTYPE WILL BE BUILT & TESTED. The first prototype will consist of an autonomous robot and a docking station. The robot will likely be based off of the platform used in ME 375. IR detectors will be installed on the sides of the robot. Additional detectors, as well as emitters, will be placed on the front of the robot. The docking station will consist of an array of IR emitting LEDs. These will serve as a beacon for the robot. Once the robot finds the docking station, some sort of interface will need to be formed. This may be done using a magnetic connector. When the robot and docking station recognize that they are connected, an arm connected to the docking station will remove the battery pack from the robot and place it on a guide that leads to the charger. This arm will be made out of laser-cut plywood or plastic, and will be powered by servo motors. A mechanism will need to be made to ensure a secure connection between the spent battery and the charger. While this process is happening, the arm will grab a charged battery and carefully place it in the robot. When the robot recognizes the battery, it will undock and continue on with its job. Auxiliary power may need to be supplied by the interface while the battery is removed from the robot. Each step in this process can be tested individually, so it is likely that we will have several prototypes being worked on at once. (ie interface, docking crane, robot control, robot sensor arrays, docking emitter, docking battery charger). The electronic components needed can all be bought, and the machined parts will be designed such that they can be manufactured in the machine shop.
IMPACT: HOW WILL THE SOLUTION CREATE BUSINESS AND/OR SOCIETY VALUE? The solution, if made applicable to robots in industry, has potential to increase productivity of machines. Reducing the turnaround time will mean that companies that utilize the robots can invest in fewer robots to perform the same tasks. This saves them money and time. Furthermore, a successful demonstration of this function could mean that autonomous robots enter new industries. For example, farmers could utilize drones for surveillance or pesticides. This would be impractical with current technology given that drone batteries last for about thirty minutes. Save the file with the following nomenclature: project name - point of contact full name.docx Email completed form to Professor Jensen at jensen23@purdue.edu to request project approval.