Oxbo is introducing AutoHarvest, a new integrated machine-learning system that automatically optimizes blueberry harvester settings in real time as field and crop conditions change. Available on MY2027 Oxbo blueberry harvesters, the technology combines onboard cameras with algorithms that continuously analyse harvesting conditions, shifting part of the machine-setting process from the operator to the harvester itself. Rather than repeatedly adjusting individual parameters, the operator or field manager defines the desired harvesting outcome through two input sliders and AutoHarvest manages the machine around those objectives.
The system can continuously adjust ground speed, harvesting-head speed, head pinch, belt speed and fan speed, allowing the machine to react as conditions vary across the field. Importantly, the optimization target can also change according to the production model: Oxbo says AutoHarvest can be configured around maintaining fruit quality for the fresh market or maximizing fruit recovery for processing. This makes the technology more than a conventional automatic setting function, because the machine is being asked to optimize several interacting parameters against an operator-defined harvesting objective.
Strategically, AutoHarvest represents another step in Oxbo’s transition from specialty harvesting hardware toward increasingly automated harvesting systems. The company has already introduced technologies including AutoFill, designed to automate container filling and reduce labor requirements, alongside EvenFill, SoftSurface and FleetCommand. AutoHarvest extends automation deeper into the harvesting process itself: instead of automating an individual downstream operation, it continuously manages how the crop is harvested. In specialty crops — where machine settings, crop variability, fruit quality and operator experience can directly affect marketable yield — transferring more of that decision-making from the operator to the machine could have particularly significant economic value.
Bottom Line
AutoHarvest signals a potentially important evolution from automated functions toward self-optimizing specialty harvesters. The strategic value is not simply that Oxbo is applying machine learning, but that cameras, algorithms and machine controls are being integrated into a closed decision loop capable of continuously adapting the harvesting process to a defined commercial objective. If this architecture expands across crops and Oxbo’s broader specialty harvesting portfolio, the longer-term opportunity is a machine that increasingly manages harvest quality, recovery and productivity as an integrated system rather than as separate operator-controlled settings.
