Before you begin
Starting knowledge
- Basic programming and a clear experimental interest.
- Distinguish a learning project from an everyday consumer appliance.
Equipment & environment
Begin entirely in simulation or recorded data. Any later hardware work needs a suitable research setup and supervision.
No fixed completion time: use the evidence gates below to decide when to advance. Resource links open the botstest directory entry; check its publisher documentation for the exact version and configuration.
Beginner
Break a household task into observable skills
Understand why everyday chores are technically demanding.
Learn in this order
- Object state, clutter and task completion
- Perception versus physical manipulation
- Simulation abstractions and unmodeled household conditions
Put it into practice
- Choose one small household-like task and define its initial conditions.
- List perception and action assumptions that a real home would violate.
A bounded task description and assumptions ledger.
Ready for the next step when…
Resources for this stage
Study household-activity evaluation definitions without assuming that simulation proves home deployment readiness.
Use an interactive simulated environment for a bounded embodied-perception or instruction task.
Examine instruction-following evaluation tasks and distinguish benchmark success from physical-home autonomy.
Intermediate
Build and evaluate a simulated skill
Create a repeatable baseline before adding language or learning complexity.
Learn in this order
- Task states, actions and observations
- Demonstrations and baseline policies
- Object/scene variation and intervention accounting
Put it into practice
- Implement or reproduce one supported manipulation example.
- Evaluate it across a declared small set of start states and scene changes.
A simulation baseline and a failure-mode comparison.
Ready for the next step when…
Resources for this stage
Build repeatable manipulation tasks in a simulated benchmark environment.
Follow robot-data recording, dataset, policy and evaluation workflows for supported hardware and software.
Model physical interactions and test control or learning behavior under explicit simulation assumptions.
Professional
Decide whether a physical research project is justified
Translate experimental results into hardware and staffing requirements.
Learn in this order
- Workspace, sensing and calibration
- Research interfaces and supervision workload
- Consumer utility versus experimental capability
Put it into practice
- Create an experiment-to-hardware requirement matrix.
- Compare a physical research platform with continuing in simulation.
A research feasibility dossier and a scoped next experiment.
Ready for the next step when…
Resources for this stage
Study mobile manipulation research and the required development and supervision workload.
Investigate an embodied-development platform and its exact control interfaces and supervision needs.
Explore a small teleoperation arm; confirm assembly, calibration and exact hardware revision support.
Common pitfalls
- Presenting benchmark success as proof of household reliability.
- Buying a research platform expecting an unattended chore appliance.
Where to go next
Move to the manipulation or imitation-learning route if the goal remains research, or use the household purchasing routes for everyday utility.
Explore other home paths →