AGD Intelligence

Ripeness/firmness grading of avocados and bananas by gentle palpation

Avocados and bananas are graded and sorted by ripeness before packing for retail/foodservice, a judgment traditionally made by hand-feel since external color alone is unreliable. The robot would gently palpate or compliantly grasp each fruit to assess firmness and select/route it accordingly, then place it without bruising into the correct pack. The fruit is fragile, bruise-prone, and ripeness varies continuously across a lot. It is hard for a robot because the discriminating signal is a controlled, sub-bruise-threshold compression force read against deformation, then a damage-free transfer. Fresh Del Monte operates a dedicated avocado packing facility in Mexico and tracks avocado quality data end-to-end, indicating active interest in produce quality/grading. We identified this through our own research; we have not confirmed the specifics with the customer directly. This page is our researched read — a starting point for that conversation.

Readiness
build now
Demand
promising
Source
researched
Failure tol.
medium
Tactile value
very high
i

What the task is

RESEARCHED · our reconstruction

Avocados and bananas are graded and sorted by ripeness before packing for retail/foodservice, a judgment traditionally made by hand-feel since external color alone is unreliable. The robot would gently palpate or compliantly grasp each fruit to assess firmness and select/route it accordingly, then place it without bruising into the correct pack. The fruit is fragile, bruise-prone, and ripeness varies continuously across a lot. It is hard for a robot because the discriminating signal is a controlled, sub-bruise-threshold compression force read against deformation, then a damage-free transfer. Fresh Del Monte operates a dedicated avocado packing facility in Mexico and tracks avocado quality data end-to-end, indicating active interest in produce quality/grading.

To confirm with the customer

Is this the actual task and sequence? What are the real tolerances, cycle rate, and reject criteria, and which steps are today's manual bottleneck? Answering these is what turns this from a researched signal into a validated use case.