Question 1Russell and Norvig define rational agents as follows:For each possible percept sequence, a rational agent should select anaction that is expected to maximise its performance measure, given theevidence provided by the percept sequence and whatever built-inknowledge the agent has.Answer subquestions a), b) and c) with reference to the example applicationslisted below, justifying your answers:A. A ‘light show’ agent that accompanies a live musical performanceby flashing different coloured lights according to the pitch, volumeand timbre of the sounds.B. A ‘virtual mechanic’ that assists a car owner in diagnosing faultsand advising how to fix them or whether to call in a professional.C. A healthcare assistant that attends to a hospital patient by e.g.providing medication at appropriate times, carrying out routine testssuch as blood pressure or temperature, and respondingappropriately to requests from the patient.a) For each application A, B and Ci. Explain what type of agent would be most appropriate:• Simple reflex agent• Model-based reflex agent• Goal-based agent or• Utility-based agent.[4 marks]ii. Explain what kinds of actions the agent would need to executeand what kinds of percepts are likely to be available, andpropose a performance measure.[6 marks]b) For each application A, B and C, explain whether the agent’senvironment is:• deterministic or stochastic• fully or partially observable• static or dynamic• single or multi-agent.[8 marks]Page 3 of 10c) For each application A, B and C, classify the agent according to thefollowing dimensions in Poole and Mackworth’s design space:• modularity: flat, modular or hierarchical• planning horizon: non-planning agent or one with afinite, indefinite or infinite horizon?[7 mark