I am wrapping up an engineering project on “getting around an indoor space by modelling human navigation behaviour” and I need help finishing the core pieces. The goal is to extract the latest findings from academic journals on how people move through buildings—specifically how they avoid obstacles in office layouts—and turn those insights into a working simulation that can be stress-tested on randomly generated floor plans. Here’s how I see the work flowing: • Literature synthesis: comb through peer-reviewed papers, identify the dominant theories and quantitative parameters that govern obstacle avoidance in offices, and summarise them in a concise technical review. • Behavioural model: translate the selected research into an algorithmic model (agent-based, continuous steering forces, or a hybrid if justified) that captures obstacle avoidance while still allowing for pathfinding on an abstract office graph. • Simulation engine: code a lightweight simulator (Python with Pygame, Unity C#, or another well-documented toolchain) that can import or procedurally generate office maps containing static and dynamic obstacles, drop an agent, and run repeatable experiments. Each run should randomise the layout so the model’s robustness is clear. • Engineering report: compile methodology, model equations, simulation screenshots, performance metrics, and critical discussion into a structured report suitable for an academic or industry audience. Acceptance criteria 1. Literature section cites only academic journals (IEEE, ACM, Elsevier, Springer, etc.) and covers at least 4 primary studies from the last decade. 2. Model code is clean, version-controlled, and accompanied by a README with setup instructions. 3. Simulation must allow a seed value to guarantee reproducibility of the random office layouts. 4. Simple summary report: Explain theory, link findings, model assumptions, simulation results, and improvement suggestions. If you have prior work on human movement, robotics path planning, or agent-based crowd simulations, this will be a perfect match.