Summer 2022 - Technical Term Scope

Owned by Austin Tailon Huang
Last updated: Jan 16, 2023
4 min read

Drivetrain

  • Source a new battery

  • Bring drive train to “M1 (milestone 1)” stage

    • fully functional prototype with controller (ebox, battery, comms equipment, gimbal, arm/science mounting plates, e-stop)

    • wired power, while looking for new batteries

    • wired cabling for controller

  • Complete FW to commutate drivetrain motors with odrives

  • Develop code to allow for drivetrain operation with physical controller (steering, driving speed modes)

  • Comprehensively test the following attributes

    • Top speed / speed verification

    • Physical constraints (size and weight, module specific mass)

    • Threshold/terrain traversal capabilities

    • Motor performance / battery life

    • Gimbal vision stability

  • Identify areas for improvement (hardware, fw, sw, controls)

  •  

  • Verify comms to fully “detach” drivetrain (no wired power or wired controller)

  • Develop code to log individual RPM/current logging for each drivetrain motor (csv format or similar)

  • Create SW behaviours to detect extended periods of motor stall, and cut power to prevent burnout

    • Motor stalling safetys - detect wheel stall and cut power to prevent permanent motor damage

    • Speed control modes (i.e. top speed, half speed, etc)

      • Different speeds may serve different functions. Full speed for traversal, half speed for fine readjustments, etc. We can learn more about this as we test/evolve with the drivetrain

  • Test on various terrain and record/detail poor mobility situations. Being developing operator, SW or hardware solutions to sort these out

    • I.e. on loose dirt and sand when going up hill, wheels loose traction and free spin without rover moving

    • Rover gets “beached”; some object gets lodged on drivetrain links - how can we get out?

  • Develop comprehensive and repeatable testing plans for:

    • Size constraints

    • Weight constraints (IRL vs CAD)

    • 1m drop test (scale up, test confidence)

    • Top speed test

    • Threshold test

    • Terrain testing (sand vs dirt vs pebbles vs whatever)

    • Upward slope testing (what limit can we hit / can we hit our design intent)

    • General traversal capabilities

    • Turn in place

    • Battery life for peak drivetrain usage

  • Start working on mechanical redesigns to address:

    • Weight reductions

    • Part count reductions / improved serviceability

    • Addressing any mechanical failures that arise during testing

    • Wheel design investigation - compliant wheels, treaded tires, etc???

  • Max range wireless drivetrain control

Gimbal

  • Determine if gimbal provides adequate 360deg vision when operating rover (subjective)

  • determine if additional static cams on (but not limited to) the following locations are required

    • bottom of rover

    • corners/edges

    • back/front of rover

  • Optimize viewing angle of wrist camera / downsize wrist camera if possible

Ebox

  • First implementation will have everything kind of thrown in there. Once we are comfortable with what goes in the ebox, redesign layout/frame to most effectively use space and allow accessibility to key connections

  • Hot weather jetson testing - making sure jetson does not throttle under peak load (autonomy or IK?) when inside the ebox

Arm

  • Bring arm to “m1 (milestone 1)” stage

    • Completely finish hardware assembly for each individual joint

      • Measure mass of each joint

      • Verify joint controls / positional control resolution

      • Independently test each joint for PID control, torque/speed output, endstop functionality, holding torque, accuracy/backlash

  • Bring end effector to “M1 (milestone 1)” stage

    • Fully assembled end effector (parallel to point 1)

      • Determine gripping capabilities (various objects, rocks, tools, 5kg weights, ensure gripper is capable of holding said items within its motors operating range)

      • Begin determining if we can detect if we are holding something (stall current? do we need an additional sensor? etc)

  • Finalize arm odrive enclosure design package

  •  

  • Fully assemble arm/implement IK

  • Play with wrist camera to determine optimal viewing angle

    • Reduce camera size if possible

  • Test workspace on static platform; verify loading capabilities

  • Determine precision/accuracy, get lots of pick and place training for operators

  • Finish allen key tool, create SW behaviour to pick up/stow tool or find alternate solution

  • Reduce arm weight with mechanical redesigns

  • Investigate controls with a dedicated operator (whats easy to use? xbox, space mouse, etc)

  • Implement proper wire strain relief hardware (make sure we arent tangling up during operations)

  • Mount arm on full rover - test COM / tipping capabilties at full loading / extension

  • Create workspace limitations in SW for mounted arm to prevent bad edge cases / collisions

  • Optimize odrive enclsoure for lower space claim/easier modularity. First implementation is going to be hacky.

Science

  • Fully assemble all hardware for science module (mechanical + electrical assembly)

  • Science chemical testing (in parallel to system assembly)

    • Performing chemical tests in labs on soil samples to create a baseline for expected results

  • Push code to independently drive each science module system (low level control hardware, just on computer or motor drivers is fine)

  • Verify functionality of each science module

    • Drill/vacuum system’s dirt collection capabilities (ideal power output for vacuum, catch points for dirt, augur redesigns, etc.)

    • Sensor read head calibration (compare sensor results to known tested values to determine if sensors for CO2, moisture, methane, etc. are reliable)

    • Elevator control functionality

    • Dino cam stability

    • Dirt to test tube delivery module (and test tube rack design)

  •  

  • After modules are verified, test/optimize full sample collection processes

  • Multiple repeated tests to figure out potential risks of false positives/negatives for chemical/ninhydrin tests

  • Study more rocks lol

Autonomy

  • Overall we are pretty behind on autonomy, so we should keep our goals this term simple

  • Autonomy mission strategy - what pathing/behaviours should our rover use when it reaches goal posts to identify QR codes?

  • Static camera QR code identification - use our zed2 to identify and recognize comp QR codes and integrate with LED matrix

    • Gimbal vibration testing - will a dynamic gimbal with its current mounting system vibrate too much and negatively impact autonomy capabilities?

  • Goal post testing hardware

  • Determine our next steps

Comms

  • We are also pretty behind on comms. We should start slow this term to understand where we are. I dont know much about comms so hopefully someone can fill the gap here

  • Get comms operational - turn on base station and small mobile antenna platform. Carry it around on a bike lol, and see if we can reach full range and when comms drop

    • Map out a small area with varying terrain, maybe a local park or forested area. Keep track of where comms drops and where we can keep communication

  • Determine our next steps