Overview and Background

ECOPRESS was developed as part of a collaborative course project aimed for the Global Leadership course at addressing the United Nations Sustainable Development Goals, specifically those related to sustainable cities, clean energy, and road safety. Our team was challenged to design a data driven, real world solution, leading to the development of a validated theoretical system and subsequent semi-mechanically prototype.

Concept Validation & Mechanical Prototyping

ECOPRESS is a roadway integrated energy harvesting system designed to recover energy lost during vehicle braking while simultaneously improving traffic safety at intersections. The project was intentionally developed in two phases:
(1) a data driven theoretical validation phase, and
(2) a mechanically realistic prototype focused on physical feasibility and power conversion.

Phase I — Theoretical System Validation (Data & Simulation-Based)

The first phase of ECOPRESS focused on validating the technical feasibility, energy impact, and safety potential of the system using quantitative analysis and simulation. This phase was supported by a full written report and presentation board and emphasized measurable performance targets rather than hardware constraints.

Key findings and targets from this phase include:

• Energy Loss Context:
  Vehicle braking in the U.S. was estimated to waste approximately 150 trillion joules of energy per year, equivalent to ~11.36 billion gallons of gasoline and ~227 billion pounds of CO₂ emissions annually.

• Kinetic Energy Recovery Target:
  Simulations predicted 10-15% recovery of vehicle kinetic energy per compression event.
  For a midsize vehicle, this corresponded to approximately ~567 J per plate, with higher outputs for heavier vehicles.

• Vehicle Speed Reduction:
  A 15-meter ECOPRESS panel array was modeled to reduce vehicle speed from 40 MPH to ~20 MPH, achieving:

  • 50-55% speed reduction for lighter vehicles

  • 65-85% speed reduction for heavier vehicles

• Vehicle Classes Simulated:
  Eight vehicle types were analyzed, ranging from 500 lb motorcycles to 100,000 lb trucks, ensuring scalability across real traffic conditions.

• Solar Energy Contribution:
  Solar panels integrated into the roadway were modeled at ~20% efficiency, serving as a supplemental energy source alongside kinetic harvesting.

• Urban Impact Projection:
  A single busy intersection outfitted with ECOPRESS panels was estimated to generate up to ~214,000 kWh per year, enough to support public infrastructure such as lighting, charging stations, or micro-mobility systems.
  
  

This phase intentionally used idealized mechanisms to establish upper-bound performance, system relevance, and real-world impact before committing to mechanical design constraints.

Phase II — Physical Mechanical Prototype (Rack and Pinion System)

Building on the performance targets established in Phase I, the second phase of ECOPRESS focused on developing a small scale, semi-physical prototype to demonstrate the mechanical and electrical feasibility of the energy conversion concept rather than a full roadway ready system. The prototype was intentionally compact, approximately 7x7 inches, and designed to validate motion transfer, power generation, and system integration at a conceptual level.

In this prototype, a spring supported top plate is manually compressed to simulate vehicle loading. The resulting linear motion drives a rack-and-pinion gear system, which converts vertical displacement into rotational motion. The pinion is mechanically coupled to a DC motor used as a generator, allowing the prototype to demonstrate repeatable linear-to-rotational energy conversion within clear kinematic constraints.

Since both compression and releases cause the pinion to rotate in opposite directions, the electrical output alternates polarity. A full bridge rectifier is therefore used to convert this bidirectional, Alternating Current signal into a consistent DC output, enabling energy capture from both the downward compression and upward spring return. An Arduino Pro Micro is integrated to monitor generated voltage and system response, providing a foundation for future control, regulation, or data logging.

A small solar panel is mounted on the top surface to represent the system’s hybrid energy-harvesting approach. While not intended to match real-world power levels, this semi-physical prototype serves as a proof-of-concept platform, bridging theoretical analysis and future large-scale mechanical development by validating force paths, energy flow, and subsystem integration.

Spider-Scan Animation

Project Overview and Background

Too(th) Choppeddd was a semester-long team engineering project for ME2110 in which I helped design, build, and test a fully autonomous robotic system capable of performing multiple coordinated tasks in a constrained, time critical environment. The project simulated a real-world scenario where a single robotic platform must navigate an environment, interact with mechanical systems, sort objects, and complete sequential objectives without human intervention.

The robot was required to operate completely autonomously after activation, complete all actions within a 40-second runtime, fit inside a compact 12 in × 24 in × 18 in starting volume, and remain under a strict $120 budget. These constraints forced careful tradeoffs between mechanical complexity, reliability, speed, and integration.

Functional Objectives

The operating environment consisted of a central work area surrounded by multiple task stations. Once activated, the robot had to:

• Translate itself out of a confined starting zone to access the workspace

• Reach and actuate a high-mounted mechanical lever (~42 in above ground), simulating interaction with infrastructure

• Collect, sort, and transport fish and eels (objects) of different sizes, placing them into correct locations while avoiding penalties

• Precisely deposit sheep (payloads) into rotating receptacles, requiring timing, alignment, and repeatability

• Manipulate and reposition multiple torches (objects) toward a target zone
  
  

All tasks had to be completed without resetting, re-aiming, or human correction, emphasizing robust mechanical design and reliable sequencing.

Requirements-Driven Design & Quantified Targets

We translated customer and competition requirements into engineering targets using a House of Quality, which guided every major design decision. Key quantified goals included:

• Total runtime: ≤ 40 seconds

• Vertical reach: ≥ 42 inches (final mechanism reached ~47-49 inches)

• Minimum linear displacement: ≥ 24 inches

• Linear extension speed: ~36 inches in under 3 seconds

• Total system cost: $113.58 (below the $120 limit)
  
  

To manage complexity, the system was decomposed into five integrated subsystems, each designed to share actuators and motion paths where possible to reduce weight, cost, and control complexity.

Mechanical System Architecture

The Too(th) Choppeddd robot was designed around a highly integrated mechanical architecture that prioritized speed, compactness, and actuator efficiency under strict size, cost, and time constraints. The final system consisted of five tightly coordinated subsystems, deliberately engineered to share motion paths, actuators, and structural elements to maximize capability while minimizing complexity.

Rather than treating each task as an isolated mechanism, the design emphasized multi-function components and mechanical reuse, allowing the robot to perform a wide range of actions within a 12 in × 24 in × 18 in starting volume and a 40-second autonomous runtime.

Integrated Drivetrain & Retraction System (Architecture Highlight)

The drivetrain was the core architectural innovation of the robot, combining locomotion and mechanism retraction into a single, integrated system powered by only two motors total for the entire robot.

This system merged what would traditionally be separate subsystems into one coordinated mechanical pathway:

• Forward translation of the robot out of the starting zone

• Retraction of extended front-end mechanisms after task completion

• Stabilization of the robot’s center of mass during dynamic motion
  
  

The drivetrain used a two-stage power flow architecture:

1. A spool-driven cable system managed linear extension and retraction of front-end mechanisms.

2. A bevel-gear-driven wheel axle handled forward motion across the arena.

By sequencing these stages mechanically rather than electrically, the robot avoided additional motors, sensors, and control logic. One-directional gearing and ratcheting elements ensured that energy was transmitted efficiently in the intended direction while preventing back-driving or unintended motion. This approach reduced system weight, wiring complexity, and cost, while improving predictability during autonomous operation.

This actuator-sharing strategy was a major contributor to the robot staying under the $113.58 total budget while still meeting aggressive performance targets.

Linear Extension & Deployment System

A front-mounted linear rail system enabled rapid deployment of task-specific mechanisms into the workspace. The system was designed to achieve approximately 36 inches of linear travel in under 3 seconds, allowing the robot to access multiple task zones early in the 40-second run.

The rail system served as a structural backbone for several subsystems, supporting:

• The high-reach lever actuation mechanism

• The object sorting and handling components

• The torch manipulation arms

By centralizing deployment through a single linear axis, the robot minimized spatial interference between subsystems and ensured repeatable alignment during each run.

High-Reach Lever Actuation Mechanism

To interact with elevated infrastructure, the robot incorporated a four-stage telescoping mechanism capable of reaching approximately 47-49 inches above ground, exceeding the minimum 42-inch vertical reach requirement.

The mechanism was driven by a cable-and-spool system, allowing compact storage when retracted and smooth extension when deployed. This architecture provided:

• High reach-to-volume efficiency

• Controlled extension speed

• Minimal lateral footprint during operation

The telescoping structure enabled precise interaction with a mechanical lever using a relatively small base footprint, demonstrating effective use of vertical space within strict volume constraints.

Fish and Eel Sorting & Handling Subsystem

The sorting subsystem was designed to autonomously differentiate between Fish and eels based on size using purely mechanical geometry, avoiding reliance on sensors or vision systems. Two arms were separated using a ~0.4-inch height offset, allowing smaller items to pass through while larger items were mechanically blocked and redirected.

This geometry-based approach provided:

• High repeatability

• Fast operation without sensing delays

• Robust performance within a moving environment

The subsystem was mounted along the linear deployment path, allowing it to engage with objects early and efficiently during the run.

Sheep Placement Mechanism

The sheep mechanism was designed to place sheep into specific sections of a rotating paddock with controlled timing and repeatability. Sheep are stored on a spring  loaded platform and guided by a vertical linear rail that constrains roll, pitch, and yaw while allowing only vertical motion.

A limit switch detects paddock alignment and triggers a single dispensing event, with a short timed delay to prevent double placement. The mechanism is compact, mechanically constrained, and integrated to operate reliably within the robot’s 40-second autonomous run.

Iterative Prototyping & Engineering Challenges

This project evolved through three major design sprints (V1 → V2 → V3/3.5), each revealing new mechanical and systems-level challenges. Early failures included insufficient actuation force, structural compliance leading to misalignment, drivetrain instability due to mass imbalance, and stress concentrations causing component failure.

Key engineering improvements included:

• Redesigning the drivetrain to eliminate underpowered pneumatics

• Adding structural reinforcements and flanges to prevent gear hub failure

• Redistributing mass to reduce overturning moments

• Increasing structural stiffness to reduce up to 6 inches of lateral deflection that initially limited success rates

By the final sprint, all major subsystems met their performance targets during testing, and system-level reliability improved significantly.

Final Performance & Results

The final system met all primary engineering requirements and operated within the defined constraints. While last-minute mechanical slip reduced final-day scoring consistency, the robot:

• Stayed under budget at $113.58

• Met all size, time, and reach constraints

• Demonstrated successful multi-task autonomy

• Placed 21st out of 72 teams (29th percentile)

• Was recognized for its highly unique mechanical architecture and aggressive subsystem integration

Formal risk analysis using Design FMEA placed the highest residual risk at 20, near the bottom of the moderate-risk category after mitigation.

Context & Disclaimer

This work was completed as part of Georgia Tech’s EcoCAR Collegiate Competition Team, in collaboration with General Motors.

DISCLAIMER:
Due to the use of confidential and proprietary General Motors information, certain aspects of this project cannot be publicly shared (Such as CAD, CAD Drawings, FEA visuals, Vehicle Tests, Certain areas of the vehicle). All information presented here: system descriptions and performance data, has been reviewed with program advisors and approved for public showcase.

The vehicle platform used was a 2023 Cadillac LYRIQ configuration that is unreleased to the public and provided exclusively to academic institutions.

Project Overview

This project was completed as part of Georgia Tech’s EcoCAR Collegiate Competition, where I worked on the Front Electric Drive Unit (FEDU) subframe redesign and motor mounting system for a 2023 Cadillac LYRIQ configuration that is not publicly available and used exclusively by universities and research institutions. The work focused on integrating the front electric motor into a custom subframe architecture while improving structural load management, vibration isolation, and overall system durability.

Unlike traditional electrical vehicles, the LYRIQ’s front subframe was not originally designed to support an electric drive unit. Our team therefore worked with a custom subframe redesign and developed new FEDU subframe mounts capable of managing electric motor torque loads while reducing vibration transmission to the vehicle chassis.

What Are FEDU Subframe Mounts? + The Redesign

The FEDU subframe mounts are structural interfaces that secure the Front Electric Drive Unit to the vehicle’s subframe and chassis. These mounts serve three critical functions:

• Rigidly locating the FEDU to maintain drivetrain geometry

• Managing torque-reaction and road-induced loads during acceleration, braking, and impacts

• Isolating motor-induced vibration to improve ride quality and cabin comfort

Our redesign went beyond updating the FEDU mounts themselves and required a coordinated redesign of the front subframe (Which was all done through Siemens NX) to properly support the new bushing-isolated architecture. Because the original subframe geometry was not designed to accommodate elastomeric bushings or the revised load paths introduced by an electric drive unit, new mounting brackets, attachment mounts, and front subframe redesigns were developed. These changes ensured that loads from the FEDU could be transferred cleanly into the chassis while maintaining proper packaging, serviceability, and structural integrity.

The updated mounting system replaces the original metal-to-metal interfaces with elastomer-isolated mounts, while the redesigned subframe provides the necessary stiffness, bracket support, and alignment to make this approach viable. Together, the mount and subframe redesign allow controlled compliance for vibration isolation while preserving precise motor positioning under torque and road impact loads. This integrated approach shifts the system from a purely rigid solution to a balanced vehicle design that improves NVH performance and long-term durability without compromising structural integrity.
dd

Analysis, Testing & Performance Targets

Structural validation was performed using Finite Element Analysis (FEA) in Altair Inspire (system-level) and Altair HyperWorks (component-level).

Load cases evaluated included:

• Torque-reaction loads: ~4,250-4,400 N

• Vertical road-impact loads: ~12-15 kN

• Acceleration loading: 20g horizontal

• Vertical loading: 8g vertical
  
  

Design targets and outcomes included:

• Targeted mount stiffness range of 19 kN/mm

• 4-5% reduction in transmitted motor vibration amplitude compared to the previous solid mount design

• Improved load distribution across the redesigned subframe

• Increased durability margin under combined torque and impact loading

System-level simulations included the FEDU, subframe, and all mounting locations together, ensuring realistic interaction between components rather than isolated part behavior.

Vehicle Integration & Packaging Coordination

Beyond the component level design, this project required full front end integration planning to ensure the new FEDU subframe mounts could be installed within the existing vehicle architecture. To accommodate the redesigned mounts and subframe geometry, several front-end systems including portions of the coolant routing, electrical wiring harnesses, and surrounding packaging hardware were temporarily removed or repositioned to create sufficient installation clearance.

This integration process involved coordinating mount placement with existing subsystems, ensuring that new load paths did not interfere with serviceability, safety, or component accessibility. After installation, all displaced systems were reinstalled and revalidated within the updated layout, preserving overall vehicle functionality while having an improved mounting solution.

This integration approach ensured that the FEDU mounting redesign was not treated as an isolated component change, but as part of a full vehicle-level system, accounting for real packaging constraints and cross-subsystem interactions typical of production EV platforms.

Work In Progress