Despite the challenges brought on by the pandemic, RoboRacing has continued to move forward on its goals and invest in the future of the team. Both the Electric Vehicle Grand Prix (EVGP; normally held in April) and the International Autonomous Robot Racing Competition (IARRC; normally held in July) were canceled in 2020. Neither is canceled yet for 2021, but even if the 2021 competitions do not materialize, the team has substantial achievements to be proud of.

One unique opportunity that RoboRacing has stepped up to is working with the EVGP officials to help make close, wheel-to-wheel racing a more viable aspect of the event. Two topics in which we have lended our expertise are (1) planning for a standardized emergency stopping system and (2) making fast localization/tracking of opponent vehicles more achievable for all teams through standardized identifying markings. While this is a nice way to support the competitive robotics community, the primary goal for the year has been getting Rigatoni, our hopeful entry to EVGP, up and running.

For the hardware subteam, the top priority has been the major mechanical systems of Rigatoni. The vehicle started life as a race-ready electric go-kart (per competition rules) which we are retrofitting with actuated steering, brakes, and a fail-safe emergency brake. While it doesn’t have a ride-along seat like Bigolii did, Rigatoni is shaping up to be a showpiece for RoboJackets.

This year, RoboRacing has finished assembling all three of these critical subsystems. In particular, the hydraulic main brakes underwent a design iteration in 2020. According to Hardware Subteam Lead Sam Morstein, the braking system “uses a stepper motor to apply force to the [hydraulic] cylinder of the brake, so that the amount of braking force can be varied.” This is the first hydraulic braking system (and hopefully the most effective) in RoboRacing team history.

In conjunction with the mechanical subteam, the electrical subteam has also made substantial progress on implementing the major systems necessary to get Rigatoni driving. The wiring for Rigatoni was revised to improve safety. The system uses more separate PCBs for individual functions, so that a failure in one board does not necessarily bring down all functions of the vehicle. A new feature of Rigatoni’s electronics this year is power over Ethernet (PoE), and the PoE daughter boards have been assembled, verified, and revised this fall. Separate control boards operate stepper motor drivers for both the steering and main braking actuation. The emergency stop board has already been verified to work on-car, receiving signals from a remote and communicating with Rigatoni’s other systems. 

New member projects on the electrical team also show a lot of promise. New member teams are working on making a custom axle encoder in-house to improve the way we measure Rigatoni’s speed and setting up multiple cameras to allow 360-degree vision around the vehicle. Another project involves a board which will allow external sensor modules to interface with the car’s existing systems. The first planned usage of this is a radar sensor facing backwards on Rigatoni to cover a blind spot when the vehicle is reversing or another vehicle is following. Between the mechanical and electrical subteam progress, the Rigatoni platform is approaching the day it can be tested with autonomy software in early 2021.

Of course, even the best robot platform isn’t much use without software to run it, and RoboRacing’s software subteam also showed concrete progress this year. It should be noted that in addition to the pandemic, the Software team also faced the additional challenge of having a large portion of its leadership graduate at the end of 2019.

The few returning members have been focused on making tweaks to the software stack to boost performance on the EVGP track. These improvements include both a global path planner and global path cost in the trajectory planner. The global path can be computed after the first lap around the EVGP track, allowing faster subsequent laps. Also, additional Gazebo (a 3D robot simulator) worlds were created to test the autonomy code with multiple opponents at the same time.

New members are generally working on improving Sedanii’s performance on the IARRC Urban Challenge. This challenge involves navigating a scaled-down grid of city streets by following signs to a destination. Notably, the team has iterated on our previous algorithm for detecting the presence and direction of the arrows which tell the vehicle to make turns.

Looking forward to 2021, the team is in a good position for EVGP and IARRC this year. Despite facing significant challenges, progress on Rigatoni hasn’t slowed, and we hope to bring you updates on testing soon!

The RoboNav team saw many changes in 2020. The spring semester saw the completion of the third iteration of Jessiii, the robot designed for IGVC (read more here). It also saw the team change its name from IGVC to RoboNav to reflect a shift in focus from a specific competition to autonomous navigation problems in general. With this name change, the team also decided to participate in a new competition: the University Rover Challenge (URC). Throughout the summer and fall semester, RoboNav has been busy working on the next IGVC robot, a swerve drive robot named Swervi, and starting the transition to URC.

Swervi is a robot that the team started working on in the spring of 2020. The decision to create a swerve drive robot came from the members’ desire to explore different drivetrains and branch out from differential drive systems, the standard drive system for many years. Swerve drive refers to the ability to pivot or steer each wheel independently. The result is a robot that can move in any direction at any time, a major benefit in the IGVC obstacle course. However, this benefit comes with the added challenge of more sophisticated designs and additional technology which requires more time for prototyping and testing. While the pandemic has slowed down work in-person, the team has managed to be very productive working remotely. The mechanical team completed the bulk of the CAD for Swervi during the summer semester and held a design review to analyze the design of the swerve module used by each wheel.

In the fall semester, the mechanical team created a test rig consisting of two swerve modules to begin testing and preparing the firmware and software required to control the robot. In the meantime, the electrical team has been busy making the necessary changes to the current electronics to support the new swerve drive. The four steering motors and four wheel motors require new encoders and a different motor controller scheme to work. The team has been revamping the system to use CAN to communicate between the mbed (which receives directions from the computer) and the motor controllers (which drive the motors). Some new members have also been making general improvements to the firmware to benefit Swervi and the upcoming URC robot. The first project is adding joystick control directly to the microcontroller (mbed). This allows members to drive the robot without having to turn on the main computer. The second project is adding small laser rangefinders to the robot’s perimeter to prevent obstacle collision if the software fails to avoid it. While unlikely, an unexpected collision could cause damage to the robot and it’s many fragile components.

Speaking of obstacle detection, the software team was also busy making improvements to the IGVC software stack. The main change was upgrading the ROS (Robot Operating System) version to the latest release, Noetic Ninjemys. The other major project was implementing 3D SLAM (simultaneous localization and mapping). SLAM allows the robot to create a 3D map of the obstacles around it and figure out where it’s located in relation to them. This is beneficial because it makes it easier for the robot to go over the ramps in the IGVC course without losing its location. Since ramps change the field of view of the camera and LIDAR, state estimation (determining where the robot is in relation to other obstacles) would fail in the past due to losing view of points of reference.

Software members have been working on other projects such as inverse projection mapping and point-cloud segmentation to improve obstacle and line detection. Previously, the team projected lines seen by the camera onto the 3D map via forward projection mapping; however, this sometimes resulted in “holes” where no line data was seen by the camera and thus not projected onto the map. The inverse projection mapping fixes this problem by eliminating these holes, which is critical to ensure the robot stays on the IGVC course. The point-cloud segmentation project is a new way of categorizing obstacles (like barrels) based on their shape. The benefit of this implementation is that it can be used to detect irregularly-shaped obstacles, like rocks, which is very important for the new competition, URC.

One of the goals of working on Swervi is to prepare the team for URC in 2022. In URC, teams create rovers capable of performing realistic activities that actual Mars rovers perform. The competition is broken up into four different missions: a science mission, a retrieval/delivery mission, an equipment servicing mission, and an autonomous navigation mission. This means that the rover made for URC will need equipment to perform experiments, manipulate objects and materials, and navigate autonomously, all while traversing the rocky terrain of the Utah desert. Since this is a much harder task than IGVC, RoboNav has already started working on a rover, Walli, in anticipation of design challenges and problems that may come up during its 2-year development.

Despite the challenges posed by the pandemic, the transition to URC has gone very smoothly. An important step was to familiarize all members with the rules of the competition, so they can think about how to tackle the new restrictions and challenges. A few members gave presentations on the rules, and the team also began a partnership with ITU Rover, an established team from Istanbul Technical University, to receive advice and share design ideas. “Overall, we’re moving faster than we originally planned,” says project manager Tan Gemicioglu, “and we’ve had some very exciting progress with the current new member projects.” The new members to RoboNav have been primarily focused on projects directly related to URC so they will be prepared for going to the new competition. On the mechanical team, members worked on researching designs for rover wheels and drivetrains. They created prototypes for the most promising drivetrains, including variations on rocker-bogie and suspension systems, using Lego components and began testing their designs to determine the best choice.

A group of new members from the mechanical and electrical teams have started work on a manipulator (robotic arm) to be mounted on Walli. They researched designs used by other teams and created some rough designs for the arm. Preliminary CAD and some prototypes are in the works and are expected to be finished in January. On the software side of things, a new Github repository for the competition has been created, and the team has begun identifying new issues to be addressed for the competition. RoboNav will shift to using ROS2 for Walli, which will require updates to any reused IGVC code and new software libraries compatible with ROS2. PM Tan has been impressed with what the team has accomplished, “despite the difficulty of juggling school work and [RoboNav] amid the pandemic.”

Of course, the team’s great progress didn’t come without challenges. Software lead Matthew Hannay noted that it’s been a difficult semester for everyone. “Being unable to meet with the [software team] and solve problems on the spot hindered our ability to make big changes in the code,” he said. Delays in shipments of parts and limited in-person shop access made the build process for Swervi longer than expected. But the team’s momentum is looking good for the spring semester. To start things off, the mechanical team will hold a final design review for Swervi as a whole and begin the full manufacturing process. Electrical members will be able to hit the ground running with testing the firmware for the new swerve modules. The software team will be able to test their new code on Jessiii as soon as possible, which will make coding and debugging for Swervi easier. And all of RoboNav will continue working on prototypes and plans for competing in URC in 2022.

This semester saw a lot of progress for our RoboCup team. The international competition of the same name in which they compete is still on for 2021, but will be taking place online. 

Thanks to a new printed circuit board assembly (PCBA) sponsorship from ASEI and the hard work of our electrical subteam, an entirely new fleet of bots were made electrically ready over 6 months before competition day. Six different boards were assembled in parallel, and more complicated boards were ordered pre-assembled. The ASEI sponsorship helped acquire PCBA for 20 mTrain boards, 5 control board prototypes, and 15 control boards and 15 kicker boards for a full fleet of robots. 

The electrical subteam has also successfully prototyped ball proximity sensors – called whiskers for short – that will be placed on the front of each bot to detect the precise location of the ball relative to the bot. The primary problem this solves is the loss of precision in cases when the field camera’s view of the ball is being blocked by a robot. The sensors will also reduce the latency of the bot’s response to the ball’s location. Normally, the ball location information is received from the field cameras and passed to the central computer, which then commands the robot to move via radio. The proximity sensor’s PCB design (pictured above) is ongoing, and the team plans to complete its production in the spring semester. 

Work also continues on our in-house microcontroller known as mTrain, primarily on the firmware side. In a collaboration with RoboRacing this semester, electrical members worked to build system files that could be used with STM32-based MCUs like the mTrain in an effort to make RoboCup’s firmware stack more developer-friendly. This will also be useful to support API functionality for those devices with minimal work from future developers. Finally, to assist the mechatronics subteam, electrical members also created a device for testing the effectiveness of the robots’ kicks by measuring ball speed as it shoots through a tunnel of sensors.

This semester, the software subteam designed a new and more intuitive gameplay system. The previous system had a number of drawbacks, including a lot of hard coded logic; for example, some types of passes were defined by referencing an exact point on the field. The new system aims to eliminate these hard coded definitions by optimizing over cost functions to find an ‘optimal’ pass, shot, position, etc for the given situation. The team also wanted the new system to be easier for new members to work with, as some aspects of the system were happening in the background, like the process that assigned bots their roles on the field. This resulted in the rest of the gameplay logic being implicitly constrained in ways that were opaque to a developer that was new to the codebase. In the new gameplay system, role assignment is more explicit so members are able to see which role robots are assigned and why. These pieces of logic will also be implemented as cost functions which are optimized over the robots. 

Other major work being done by the software team includes successfully moving most of the codebase to ROS 2. This effort was facilitated in part by a successful transfer of knowledge from other RoboJackets software subteams like RoboNav and RoboRacing, which have years of experience working with ROS. Additionally, RoboCup is in the process of setting up an automated testing system to cut down on time spent on manually testing code. 

The mechatronics subteam, newly formed as of the fall semester, is already paying dividends as RoboCup has begun prototyping in-house again. Previously, all manufacturing was outsourced to ensure uniformity of the robot fleet, but this resulted in less hands-on machining experience for RoboCup mechanical members. This semester, they have been training their members on the common machining area (CMA) machines, including mills, lathes, and waterjet, and have used their new capabilities to experiment with variations on each subsystem to optimize the robots’ designs. Like the electrical team, the mechatronics subteam has been pushing hard to get the fleet ready, with five bots assembled and ready to go for testing in the spring. 

A major achievement was the subteam’s work to optimize the ratio of kicker boot to plunger. Theorizing that inconsistent behavior of the kicker was due to less magnetic material being inside of the solenoid, the team purchased a variety of plungers and boots, and after some experimentation (pictured below) found that the longest plunger gave the most consistent and fastest results. The mechatronics subteam is also in the process of troubleshooting the bots’ new dribbler design, which they found interfered with the kicker inside the robot and blocked the solenoid plunger from coming out. After some prototyping and 3D printing, the subteam lead, Alyssa Gordon, states that they have a solution but are still trying to optimize the design further.

The mechatronics subteam is also experimenting with the bots’ center of mass to eliminate some issues with motion control. The current bot design places the rear wheels further from the center of mass, resulting in those wheels getting less traction. Without an even weight distribution, the robots’ rear wheels tend to spin out when trying to move sideways, causing the robot to spin in place. By moving the center of mass backwards – currently being prototyped by adding brass plates to the back of the bots – the traction force is balanced between the front and rear wheels and prevents this loss of control. 

Despite the challenges imposed by COVID and disappointment of the competition moving online, the RoboCup team has had a successful fall semester, with almost a whole fleet ready for software testing in the spring, the new mechatronics team making valuable contributions while moving manufacturing back in-house, and each subteam progressing impactful projects that will continue into next spring.

RoboJackets has participated in autonomous and RC Sumo competitions in California and Japan in the past as part of BattleBots. This semester RoboJackets has formed a separate team to compete in these competitions: RoboWrestling! For a quick recap, autonomous Sumo competitions involve two robots going head-to-head to localize and push their opponent off of the field. These robots typically have hundreds of pounds of effective weight due to the magnets on their underside pulling down towards the steel field. Additionally, robot reaction times need to be in the sub-second range to quickly defeat their speedy opponents. RoboWrestling has taken on new projects as a newly formed independent team while they prepare for the competitions in the coming year.

The team has entered their robots Kirbi and Gucci, which are on their second and third revisions, respectively, into prior autonomous Sumo competitions. In the beginning of RoboJacket’s autonomous sumo participation, the team was focused on creating a strong mechanical and electrical foundation for the software team to develop on. Now, our robots have matured to relatively stable platforms that are incrementally improved. Different design concepts can be tested on each bot individually, and the better-performing designs can be standardized between them. Some recent changes on these robots include increasing the TFMini Time of Flight (ToF) sensor density on the front by relocating the side sensors to the front. The ToF sensor uses lasers, similar to a LiDAR, to measure the time it takes to bounce off an object. This information can be used to estimate the location of an opponent. Since it is more likely for the robot to face their opponent directly, finer granularity of position estimation in that direction is desirable. With the greater density of sensors in the front of the robot, versus originally equal distribution across the front and sides, the robots can more accurately pinpoint their opponent’s location.

Each of the subteams have been working on projects on an individual basis. The electrical subteam has focused on testing new PCB designs for Gucciii and Kirbii. The board updates  this year include the control, radio, and power distribution board. The new radio board allows for remotely configuring the robot and testing different strategies very quickly without the time-consuming process of uploading new code to the microcontroller. The other new board is the power distribution board, which takes over the previous partial job of the motor controller for powering the whole robot while also adding current sensing capabilities.

Software has been exploring new projects this semester as well. The primary focus has been developing a simulator to help quantify and visualize the robot dynamics. The primary focus is on collisions, which often make or break an autonomous Sumo competition round. The team is currently focusing on the visualization aspect, with a very early render of a match shown below. Additionally, the team has been working on developing firmware for the new radio board to incorporate remote configuration of strategies. 

The mechanical subteam has also been making progress. They have been thinking through the designs of a brand-new robot, yet to be named. This has been helmed primarily by the new members on the team, allowing them to learn from the previous designs and bring in novel new ideas. One such new idea is a magnetized, hinged plow design. The plow, used to scrape the opponent off of the competition field, adds length to the robot which could be detrimental if the opponent were to get underneath the robot. A hinged plow significantly reduces the leverage gained by the opponent’s robot by hinging at the base to reduce the length of the lever arm created by the robot’s body. There are also magnets on the plow to keep it close to the steel field, reducing the likelihood of the opponent getting under our robot in the first place. 

A great development from the formation of this new RoboJackets team has been the increase in interdisciplinary subteam projects. Electrical and mechanical have been working together to develop test beds for new sensors, along with generating motor speed and torque calculations seen below. Mechanical and software have been focused on modeling the robot dynamics in the simulator, generating the equations to accurately simulate collisions. Software and electrical have working on the firmware for the radio boards, helping to define necessary connections between the microcontroller and radio boards for communication. Overall, the team has grown and developed since becoming independent, and they are excited to see where the spring semester takes them.

Aside from the new member 2 lb program in Battlebots that we covered in our December issue, Battlebots subteams work on a variety of larger bots. This year our subteams of experienced members are working on combat robots in two weight classes: 12 lb hobbyweights and 30 lb featherweights. In this article, we’ll be sharing an update on the progress of each of these bots in the words of the subteam leads, and providing some pictures to show you what we’re working on. We also asked each subteam lead about what the biggest challenges they faced were. For any bot that’s a new iteration of an old bot, we also asked what the biggest changes were.

First, an update on Hocki, a “melty brain” hobbyweight whose entire body spins to hit its opponents with a single tooth instead of utilizing a separated drive and weapon system in more conventional combat robots. The relative speed of the two drive wheels is varied to control where the robot goes, requiring more advanced electronics and software. The team has made a lot of progress, including a complete CAD design, initial manufacturing of the chassis as well as some of the custom electronics required to operate the robot. Additionally, testing has been conducted to ensure the proper function of some of the control boards. The most recent work involves integrating the sensors on the control board in the software. 

Varun Madabushi, bot lead for Hocki, highlighted a few challenges the Hocki team is working on, including manufacturing the outer ring of the robot. The ring was originally designed to be manufactured by wire-based electrical discharge machining (EDM), but because of issues in accessing on campus EDM machines, the manufacturing was outsourced to an external machine shop. On the electronics side, they have to balance complex systems with tight size requirements, especially on the design board which requires 4 subsystems. Additionally, to control the robot, the control software has to model the dynamics of the robot with limited data inputs from sensors. The team is planning to implement an extended Kalman filter, an industry standard nonlinear estimation strategy, to estimate the angle of the robot during its rotation.

Next, we have an update on Maorii, the second version of a featherweight beater bar spinner. The team has made a great deal of progress this semester, and they have machined most of their chassis components. They plan to begin installing their electronics and drive system and machining their weapon at the beginning of next semester. Bot lead Dennis Crawford highlighted that the main challenge was determining the best way to manufacture the clamp that links their weapon pulley to the weapon itself. One success he highlighted was getting a significant number of their members trained on using the waterjet, enabling us to better utilize a very useful tool at our disposal. The main upgrades from the first version of Maori are an improved weapon electronic speed controller (ESC) to increase weapon spin up speed,an updated weapon design to lean further into the beater bar design, as well as a new armor design to better spread out the force of weapon impacts.

Next, an update on Samuraii, the second iteration of a featherweight vertical spinner with a hinged weapon mount to ensure the weapon hits consistently even when the robot is inverted. The progress the team made this semester was finalizing their CAD design and machining some essential components including major chassis plates and turning pulleys. Dylan Adriano, Samuraii Lead, highlighted the limitations on manufacturing time imposed by COVID restrictions as the biggest challenge the Samuraii team faced. The team had to work to design the robot to be easily manufactured and had to focus on planning shop time to efficiently get their manufacturing done. He highlighted the fact that they were able to overcome these challenges and still make progress through careful planning, a big success for the team. In line with these manufacturing challenges, the team focused on design for manufacture and assembly in this iteration of Samuraii, including a simplified rectangular chassis and reducing the required number of bolts for assembly. Additionally, the team added another set of wheels to improve mobility, and reinforced the chassis to improve robustness.

Our penultimate robot update is Valkyrii, the second iteration of a featherweight “undercutter” style horizontal spinner. The team made a lot of progress over the summer reworking the weapon and chassis design, and hit the ground running on machining bot components in the fall semester. They’ve made significant progress in machining chassis plates. They also are reusing the pulley from Valkyri, which will save them valuable machining time. Sam Li, Valkyrii lead, said the biggest challenge in the design process was managing the weight of the robot. At first, they were as high as 10% overweight, but they were able to modify the parts to get the robot to meet their weight requirements while balancing manufacturability. This new version of Valkyrii is designed to utilize modular weapons that can be used interchangeably to better respond to challenges imposed by an opponent. They have two different weapon designs: a disc type weapon and a bar type weapon. The disc has a high moment of inertia, and therefore hits hard to take down tough opponents. The bar has a lower moment of inertia, enabling a faster spin up time, but weaker hits. This is to counter faster robots that rely on ramming attacks, like our eternal enemy, wedges. Additionally, they changed their armor to include nylon fibers to increase strength and a curve to cause enemy blows to glance off. They also made upgrades to the electronics system to include programmable ESCs and censored motors for the drive system to improve drive control.

The final robot update this year is Chonki, a hobbyweight shell spinner with a shuffling drive train. The team has made major progress, including an almost-completed chassis assembly. The remaining work is the construction of the shuffler system and weapon manufacturing. Brian Epstein, Chonki lead, said that the main challenge for Chonki was manufacturing the outer ring. The team considered a few different designs, starting with a concept that involved bolting or welding several water jetted layers to create the stepped ring shell. The team made heavy use of FEA to evaluate their designs iterations, eventually settling on taking a single piece of steel that will be CNC milled to the specified shape. This new design presents a major challenge in handling and fixturing a 250 lb block of steel in our CNC mill.

This semester has been a productive one for Battlebots’ experienced member teams, with a variety of challenging robot designs and restrictions imposed on manufacturing. Despite these, the teams have persevered, and each team has made significant progress in manufacturing their robots. The progress they’ve made this semester will ensure the teams are set up for continued success in the spring semester.