Tags: SummerBlogs, Profiles.DiegoReyes

Diego Reyes Summer 2025

"Week Five, Six & Seven - July 21 - August 8 "

These past three weeks were spent making improvements to the end effector. I added grip tape to the anchor in order to make its surface less smooth, reducing some of the slipping from the hooks. I also added another pair of hooks to make the hooking more stable. I also added a motor with a gear in order to actuate the end effector.

"Week Four - July 14 - July 18"

This whole week was spent designing and printing a 2D version of the anchor, consisting of two inverse hooks, mounted on a central hub and moved by gears. I was also designing the mechanism to which my anchor will attach to.

"Week Three - July 7 - July 11"

• Collaborated with research mentors Venny and Stanley to brainstorm design strategies for brace manipulation, focusing on how to passively or actively anchor ReachBot’s end effector to external surfaces during dynamic or extended operations.
• Conducted a comprehensive literature review on anchoring end effectors, exploring mechanisms including:
  • Magnetic attachment systems and their limitations in planetary or non-ferrous environments.
  • Suction-based gripping and its dependence on surface geometry and vacuum integrity.
  • Mechanical hooks and latching systems, particularly in cluttered or constrained environments.
• Developed a novel concept for an anchoring end effector:
  • A tripod-like structure actuated by a single motor, enabling symmetric expansion.
  • Each beam terminates in a reverse hook that engages with an internal lip of the anchoring target.
  • Proposed an anchoring interface: a cylindrical hole with a recessed internal lip. The end effector enters in a closed configuration and expands once inside; pulling back on the boom causes the hooks to mechanically latch onto the lip, creating a passive mechanical constraint for rigid anchoring.
• Continued developing the Onshape CAD model of the anchoring end effector, focusing on:
  • Integration of the Dynamixel motor, considering mounting constraints, torque requirements, and cable routing.
  • Geometry of the beams to ensure uniform expansion.
  • Geometry of the hook that would create the best latching
  • The development of a central hub that will host a tendon system to expand the beams
• Participated in an another Arduino workshop led by Mark:
  • Gained hands-on experience with breadboard wiring, including resistors, LEDs, and pressure sensors.

Week Two - June 30 - July 4

• Attended an Arduino programming workshop led by Mark, where I:
  • Learned to implement non-blocking code using millis() instead of delay() functions, improving real-time responsiveness and enabling asynchronous control logic—an essential technique in embedded systems and robotics.
• Contributed to a data collection experiment for ReachBot, focusing on:
  • Establishing the correlation between encoder wheel ticks and linear boom extension length.
  • Helped calibrate the system by measuring multiple extension lengths to create a conversion model between raw encoder data and real-world displacement—supporting ReachBot’s motion planning accuracy.
• Collaborated with a team to program and control the ReachBot end effector for a bottle manipulation task:
  • Focused on developing the Python-based joint control system, initializing all actuators to neutral positions and manually tuning joint angles through iterative testing.
• Continued learning in robotics fundamentals, including:
  • Forward and inverse kinematics.
  • Kinematic modeling of revolute and prismatic joints in 2D space.
  • Introduction to the Jacobian matrix and its role in mapping joint velocities to end effector motion, critical for control and manipulation tasks.
• Assisted in data collection for a tree-perching aerial drone project:
  • Captured images of diverse tree structures around campus to support perception algorithm development and environmental modeling.
• Conducted an in-depth review of the ReachBot architecture:
  • Focused on system integration: the perception model, microspine gripper mechanism, extendable boom architecture, and the robot’s mobility system.
  • Analyzed system-level trade-offs in payload, grasp strength, and workspace reach for planetary cave exploration missions.
  • If interested about a more in-depth summary of what I learned, here is a link!

https://docs.google.com/document/d/10ZcAOVb1SHnMBtk1lxjvz3ipsau98LJ_7yAADrIFg7g/edit?usp=sharing

• Began preliminary research into brace manipulation strategies for our upcoming ReachBot project:
  • Explored concepts in compliant grasping, anchor-based locomotion, and constrained motion planning.

Week One - June 23-27

• Designed a tetrahelix structure using Onshape, exploring the geometric and structural principles behind modular robotic design and gaining proficiency in parametric CAD modeling and assembly constraints. When designing the Tetrahedral I did not do a very good job of fileting the shape which made it really hard to fasten the parts together in the assembly studio
• Learned the fundamentals of Dynamixel XL430-W250-T actuators, including:
  • Torque-speed characteristics, control modes (position, velocity, PWM), and communication protocols
  • Explored DYNAMIXEL Wizard 2.0 for real-time monitoring and configuration—gained practical skills in actuator calibration, ID assignment, and firmware-level diagnostics.
• Granted access to the ReachBot end effector CAD model, where I:
  • Studied the mechanical integration of multiple degrees of freedom (DOF) and how joint placement and linkage design enable grasping functional.
  • Assembled the end effector hardware, mechanically mounting Dynamixel actuators, ensuring clean wire routing, structural alignment, and torque accessibility.
• Completed an in-depth literature review on agricultural robotic end effectors:
  • Investigated anthropomorphic grippers used in soft fruit and vegetable harvesting, especially for irregular objects like pumpkins.
  • Analyzed the mechanical design trade-offs between compliant mechanisms, dexterity, and grip force control.
  • If interested about a more in-depth summary of what I learned, here is a link!

https://docs.google.com/document/d/1vH3snnXikke4GkY-Seam5L0Aoe2f6KTUwtcvIA8LCZQ/edit?usp=sharing

• Learned about the integration of ReachBot’s robotic arms with the ANYmal quadruped, leading to the GIRAF (Grasping Instrument for Remote Access and Fetching) platform:
  • Learned about system-level robotics integration, including mechanical interfacing, payload limitations, and real-world deployment challenges.
• Participated in a live demonstration of GIRAF for DaVinci Camp, during which I was able to take control of the robot using its user interface
• Completed required online and in-lab safety training for Stanford’s BDML (Biomimetics and Dextrous Manipulation Lab) and gained access to lab tools, Slack workspace, and internal resources for collaboration.
• Attended the Summer Research Kickoff at Portola Redwoods State Park. I had a lot of fun hiking and talking with the other students. It was fun seeing everybody not having to stress about their projects.